Datasets:

krvhrv

/

crawl-books

like 0

axis of rotation (z-axis): (a) an ideal rectangular profile from a point source; (b) a realistic trapezoidal profile caused by the penumbra produced by a finite focus; (c) profile for a similar configuration as (b) but with a reduced collimator opening; (d) profile influenced by the increased penumbra resulting from collimators closer to the focus. Relative dose Dose rate 285 Dose rate dependence also because of the small contribution of the extra focal radiation. The width of the penumbral region is some millimetres due to the finite size of the virtual x-ray source and lateral electron disequi- 1 librium. The penumbral region has a long shallow fall off in dose due mainly to the scatter radiation and less due to transmission through the jaws. The scatter radiation increase with depth and field size and subsequently the penumbral dose in the tail increase B: Dose-rate~1 cGy/min in magnitude with depth and field size. A: Dose-rate ~ 1 Gy/min Related Articles: Penumbra, Radiation field, Beam flatness Dose rate (Nuclear Medicine) Dose rate is defined as the absorbed dose per unit time. The dose rate can be used to estimate a biological effect provided the effect is proportional to the dose rate. If an equal radi- DA DB ation dose is received over different time intervals, different biolog- Dose D ical effects are observed. Fractionated irradiations allow damaged cells to repair between irradiations (classic dose rate effect). FIGURE D.67 Cell survival curves become straighter as dose rate is Further Reading: Hall, E. J. 2000. Radiobiology for decreased. the Radiologist, 5th edn., Lippincott Williams & Wilkins, Philadelphia, PA, pp. 74–75. shouldered cell survival curves observed at high dose rate gradu- Dose rate constant ally become straighter as only lethal damage is expressed due (Nuclear Medicine) The dose rate constant, Γ is an estimation of to the repair of sub-lethal damage. This is illustrated in Figure the effective dose received from x-rays and γ-rays from external D.67 where curve A, typical of that obtained at a dose rate of sources. Γ is measured at 1 m distance and the unit is mSv m2/ around 1 Gy/min, has a marked curvature compared with curve MBq h. Γ depends on nuclide characteristics like the number of B, typical of that obtained at a dose rate of around 1 cGy/min. The x-rays and γ-rays per disintegration, their energies and the attenu- dose recovery factor (DRF) provides a measure of the amount ation properties of the tissue. Generally low energy (typically <10 of sparing associated with the reduction in dose rate and is the keV) x-rays and γ-rays are excluded from the dose rate constant ratio of the doses that give a specified surviving fraction: DB/DA calculations since they do not pose any greater hazard because of in Figure D.65. The dose needed for 1% survival is around one to their low penetration through syringe and vial walls. three times higher at 1 cGy/min than at 1 Gy/min. Further Reading: Cherry, S. R., J. A. Sorenson and M. E. Most normal tissues show considerable sparing as the dose Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, rate is reduced. For example, a study by Down et al. (1986) on Philadelphia, PA, pp. 434–435. mouse lung irradiated with Cobalt-60 γ-rays found that the DRF for acute pneumonitis, measured at an incidence level of 50%, Dose rate dependence was 2.6 when the dose rate was reduced from 100 to 2 cGy/min. It (Radiotherapy) The response of tumours and normal tissues to was not possible in this experiment to reduce the dose rate below radiation depends on the rate of dose delivery: a reduction in the 2 cGy/min but when the incomplete repair model (see article on dose rate is generally associated with a reduction in the radiation Linear quadratic (LQ) model) of Thames (1985) is fitted to the response (although a few exceptions to this rule have been noted). data, it predicts a DRF of 4.4 when the dose rate is reduced further In external beam radiotherapy, dose is typically delivered to 0.01 cGy/min (again compared with the dose rate at 100 cGy/ at dose rates of 1–5 Gy/min and the linear quadratic model has min). It is expected that the proliferation of stem cells in the lung been shown to describe the response to radiation both in vitro will produce even greater sparing at very low dose rates. and in vivo well. However, as dose rate is lowered, the time taken In some situations an ‘inverse dose-rate effect’ has been to deliver the radiation dose is extended and it becomes possible observed which has been attributed to the low dose rate allow- for the radiation response to be modified as a result of the fol- ing the progression of cells through the cell cycle into the more lowing processes: repair of sub-lethal damage, redistribution, sensitive phases and thereby suffering greater damage. A further repopulation and reoxygenation. Such considerations are required lowering of the dose rate will allow cells to escape the radiosensi- in the clinical situation of LDR brachytherapy and in targeted tive phases and this combined with the effect of cell proliferation radiotherapy. will produce a shallower survival curve. Repair is the fastest of the four processes. When exposure is Abbreviation: DRF = Dose recovery factor. on the order of its half-time (∼1 h), considerable repair will take Related Articles: Cell cycle, Cell proliferation, Cell sur- place. Therefore, repair will modify radiation effects over the vival curves, Dose response model, Linear quadratic (LQ) dose-rate range 0.001–1 Gy/min. Repopulation is a much slower model, Radiosensitivity, Redistribution, Reoxygenation, Repair, process with doubling times for human tumours and normal tis- Repopulation, Surviving fraction sues ranging from a few days to a few weeks. Hence, significant Further Readings: Down, J. D., D. F. Easton and G. G. Steel. repopulation will only occur during a radiation exposure when its 1986. Repair in the mouse lung during low dose-rate irradiation. duration is a significant fraction of a day which requires a dose Radiother. Oncol. 6: 29–42; Nias, A. H. W. 1998. An Introduction rate below 2 cGy/min, depending on the cell proliferation rate. to Radiobiology, 2nd edn., John Wiley & Sons Ltd., Chichester, Redistribution and reoxygenation will modify response over an UK; Steel, G. G. 2002. Basic Clinical Radiobiology, 3rd edn., intermediate range of dose rates. Arnold Publishers, London, UK; Thames, H. D. 1985. An incom- As the dose rate is lowered in the range from 1 Gy/min down plete-repair model for survival after fractionated and continuous to 1 cGy/min, the radiosensitivity of cells decreases and the irradiation. Int. J. Radiat. Biol. 47: 319–339. Surviving fraction Dose rate distribution 286 Dose-to-medium calculations Dose rate distribution Dose response curve (Radiotherapy) The dose rate distribution is particularly impor- (Radiation Protection) Dose response curves such as the linear tant in brachytherapy. It is a map of the dose rate around a radio- response curve, threshold dose response curve and non-linear active source. The dose rate distribution depends on radionuclide response curve, etc. are used either separately or in combination decay properties and the activity of the source. It can be used to as models to describe the response of the human body to expo- calculate the absorbed dose in different locations around one or sure to various types on ionising radiation from both internal and several point sources. external exposure, and at high and low doses and dose rates. This response may be described mathematically as the dose response function. Dose rate effectiveness factor (DREF) Related Articles: Linear no-threshold dose model, Stochastic (Radiation Protection) The dose rate effectiveness factor (DREF) effects, Non-stochastic effects is used to estimate the stochastic biological effects from expo- sures to ionising radiation that are spread over time. This allows us to account for effects of healing mechanisms within the body Dose response function D that reduce the implications of a radiation dose when applied at (Radiation Protection) The response of the human body to differ- a low dose rate. For example, an instantaneous dose of 1Sv to ing doses of ionising radiation may be described mathematically a human produces an approximate risk of 5% of a fatal cancer as the dose response function, which can then be plotted graphi- developing. But, if that dose was spread across a longer time span cally as a dose response curve. (from days to weeks or years), then the risk would be reduced by Related Articles: Dose response curve, Stochastic effects, a factor dependent on the timescale of exposure. Non-stochastic effects In general a factor of between 2 and 10 would be expected, depending on the timescale. However, for radiation protection Dose response model purposes, and to err on the side of caution, a factor of 2 is used for (Radiotherapy) Dose response models are mathematical models all circumstances. that describe radiobiological phenomena pertaining to the response Hyperlinks: EMERALD (DR module), www .emerald2 .eu of mammalian cells (tumour or normal tissue) to radiation. In gen- eral, dose response models describe how the probability or fre- Dose reference point quency of a specific response changes with the dose. Such models (Radiotherapy) This is generally the point at which the prescrip- are usually derived from the fitting of mathematical functions to tion dose is delivered. In brachytherapy, e.g. a point 0.5 cm directly data obtained either from laboratory experiments on cell cultures, out from the surface of the ovoid pair along the left to right axis animal experiments, or human data obtained either by studying of the pair is often used as the dose reference point (Figure D.68). patients who have received radiation therapy or those involved in Another example also in brachytherapy is Point ‘A’ in the radiation accident/incidents such as the victims of Hiroshima and Manchester system. In external beam radiotherapy, the isocen- Chernobyl. Examples of such models include the linear quadratic tre is most often the dose reference point for prescription (ICRU model, tumour control probability and normal tissue complication report 29). probability, Dose reference points can also be used to compare doses to Related Articles: Linear quadratic (LQ) model, Normal tissue critical organs between different prescribing systems. An exam- complication probability, Radiobiological models, Tumour con- ple of this is the International Commission on Radiation Units trol probability and Measurements (ICRU) bladder point in intra-cavitary brachy- therapy (ICRU report 38). This point is determined by taking Dose tolerance radiographs with a radio-opaque balloon in the bladder prior to (Radiotherapy) Radiation treatment inevitably affects normal tis- treatment. A point at the centre of the balloon is marked and the sue and so may cause radiation induced adverse effects. In radio- dose recorded. therapy, it is generally the case that the total dose that can be Further Readings: International Commission on Radiation tolerated depends on the volume of tissue irradiated – the dose- Units and Measurements. 1978. Dose and specification and volume effect. Additionally, the tissue architecture is thought to reporting of external beam therapy with photons and electrons be important in determining the tolerance dose for partial organ ICRU report 29. ICRU, Bethesda, MD; International Commission irradiation. For further information, see the articles on Tolerance. on Radiation Units and Measurements. 1985. Dose and volume Related Article: Tolerance specification and reporting for intra-cavitary therapy in gynaecol- ogy ICRU report 38. ICRU, Bethesda, MD. Dose-to-medium calculations (Radiotherapy) Historically, treatment planning systems (TPS) have modelled the impact of different tissue densities in terms Sup Left Right of scatter and attenuation but then reported the dose at each location as the dose-to-water (Dw), in a manner consistent with protocols for absolute dosimetry. However, Monte Carlo dose calculation engines typically calculate dose-to-medium (Dm), Ovoids pair which may or may not then be converted to Dw. Differences Inf between Dw and Dm could affect plan evaluation, dose report- Dose reference point at 0.5 cm from the ing and dose verification. Currently, there is no consensus on surface of the ovoids which of Dw or Dm might have greater biological meaning. Dw has, however, long been used in clinical studies for outcome cor- FIGURE D.68 An ovoid pair showing the position of the dose reference relations against dose. point. Related Article: Treatment planning system Dose-to-water calculations 287 Dose volume histogram Dose-to-water calculations (Radiotherapy) Either dose-to-water (Dw) or dose-to-medium (Dm) may be reported by Monte Carlo dose calculation engines. For more information,

see Dose-to-medium calculations. Related Article: Dose-to-medium calculations Dose tracking software (Diagnostic Radiology) Recently, dose recording and the report- ing requirements for medical procedures have become more rigorous. This is especially true for CT studies and interventional pro- cedures where scientific and medical institutions have introduced legislation which underpins the importance of collecting dosi- metric information. In addition to this, the importance of storing information gathered from different imaging modalities in data- D base form is now recognised. The use of ‘dose tracking software’ enables the automatic archiving of required information and various commercial and FIGURE D.69 Farmer ionisation chamber. (Courtesy of EMERALD ‘open source’ codes are available which are capable of this. project, www .emerald2 .eu) The most comprehensive of these allow: • Multi-modality data collection (CT, interventional Abbreviations: TPS = Treatment planning system, TLD = equipment, mammographic and DR systems) Thermoluminescent dosimeter, EPID = Electronic portal imaging • Multi-manufacturer data collection device and IMRT = Intensity modulated radiation therapy. • Flexible interfacing for capturing and transmitting data Related Articles: Field verification, Treatment verification, (different manufacturer and equipment dependent dose Detectors, Treatment planning system, Electronic portal imaging reporting capability exists but standardisation of dose device, Intensity modulated radiation therapy reporting has yet to be achieved). Dose volume histogram Integrated statistical analysis is possible and reports can be gener- (Radiotherapy) A dose volume histogram (DVH) is a tool used ated which are device-, operator-, protocol- and patient-specific. in radiotherapy to aid evaluation of treatment plans. It provides In addition, data can usually be exported so that more in-depth a summary of the dose distribution throughout a specifically analysis can be performed at a later stage. defined volume of interest, e.g. the target volume or an organ at Dose tracking software can effectively support the optimisa- risk (OAR). There are two alternative forms of DVHs, as shown tion process. It is also recognised that contributions from different in Figure D.70. A differential DVH shows the total volume of vox- professional groups (radiologists, radiographers, medical physi- els receiving a dose in a specified dose interval, against a set of cists, etc.) are necessary to review collected dose data, identify equally spaced dose intervals. Alternatively, a cumulative DVH opportunities, standardise and adjust protocols and compare local shows the total volume of voxels receiving a dose greater than, or and national diagnostic reference levels (DRLs). equal to, a specified dose, against dose. This is the format most widely used in radiotherapy. The volume and dose can either be Dose verification specified as absolute units, or as percentages (either of maximum (Radiotherapy) Verification is a key component of radiotherapy. or prescribed dose). The purpose is to ensure that the treatment is geometrically DVHs are helpful in assessing the dose uniformity across the and dosimetrically accurate, delivering the desired high dose to target volume, as well as evaluating if the dose received by an the target and achieving the required dose limits to normal tis- OAR is above the tolerance value. This is especially useful for sues. Dose verification may be done prior to treatment or during evaluating the dose received by a parallel organ, such as the lung treatment. Pre-treatment Dosimetric Verification: The planned treat- ment beam arrangement is transferred to a model of a phantom Volume per in the TPS, and a comparison between the prediction of the unit dose TPS and the dose measured in the phantom is made. Point dose Differential and 2D dose (or 3D dose) distribution should be verified. For dose volume the verification of the point dose ionisation chambers (Farmer histogram type or PinPoint chamber) can be used (Figure D.69). For rela- tive dose distribution films, radiochromic films, 2D dosimeter d1 arrays (2D matrix of detectors), EPID, or radiosensitive gel can Volume (%) be used. X = % of organ Cumulative receiving Dosimetric Verification during Treatment (In Vivo dose volume a dose >d1 Dosimetry): This involves measuring the dose at points where the histogram X = shaded area in beam enters the patient (usually diodes or TLD), where it exits differential DVH the patient, or in an anatomical cavity. Other approaches involve d1 measuring the dose in a plane outside the patient using a detector such as an EPID. This is called in vivo dosimetry. FIGURE D.70 Indication of how DVHs are constructed. Dose volume histogram (DVH), differential 288 Dosimeter air kerma), multiplied by the width of the radiation beam. This measure is particularly useful for narrow slit beams of radiation and is considered the most appropriate quantity for the assessment of patient dose with panoramic or orthopantomograph (OPG) den- tal units. Diagnostic reference doses for OPG units are expressed in terms of the DWP measured in mGy .m m. The DWP represents a measure of the total beam energy across the beam width independent of the exact shape of the (a) Dose (%) 100 (b) 0 Dose (%) beam profile. The DWP of a square profile can simply be calcu- lated from the absorbed dose measured within the beam, using FIGURE D.71 Examples of integral DVHs for PTV (a), and OAR (b). TLD or a small area solid state detector, multiplied by the full The solid line shows the ideal situation, whereas the dotted line is more width half maximum of the beam profile usually determined realistic. with the use of film. If the beam has a more Gaussian profile then the DWP can best be calculated from the total dose across D or liver, where dose limits to normal tissue are often expressed in the beam width as measured with an isotropically sensitive terms of the percentage volume receiving a specified dose. They detector, such as a CT pencil chamber, multiplied by the cham- can be used as a graphical way of comparing multiple treatment ber length. plans on a single graph. An ‘ideal’ cumulative DVH for a target volume and a critical volume are shown in Figure D.71. However Dosemeter they must be considered alongside the spatial dose distribution for (Radiation Protection) The specific use of the terms dosimeter and a complete picture of the treatment plan. dosemeter in English refers to the following common meaning: They are also used as input data to obtain statistical measures Dosimeter is usually a passive device used for patient or staff such as tumour control probability (TCP) and normal tissue com- dosimetry. This device accumulates the dose over a certain period plication probability (NTCP). These measures have been shown of time and is retrospectively read. Example of dosimeters are to be very sensitive to changes in the shape of the DVH, and film badges, thermoluminescent dosimeters, etc. hence, an accurate DVH calculation is important. Guidance on Dosemeter is an active device which measures absorbed dose methods for DVH calculation is given in the references. (or dose rate) in real time. Example of dosimeter is ionisation Abbreviations: DVH = Dose volume histogram differential chamber with electrometer. and DVH = Dose volume histogram integral (cumulative). The absorbed dose depends on the mass absorption coefficient Related Article: Dose area histogram of the material (medium). If the measurement of the dose Dair Further Reading: Drzymala, R. E., R. Mohan and M. S. is performed in the air then to know its value Dm in a medium Brewster. 1991. Dose volume histograms. Int. JROBP, 21: 71–77, (for the same type of radiation) we have to calculate it from the PMID: 2032898. formula: Dose volume histogram (DVH), differential D (Radiotherapy) See Dose volume histogram air x(mm /rm ) Dm = (mair /rair ) Dose volume histogram (DVH), integral cumulative where (Radiotherapy) See Dose volume histogram μm and μair are the linear absorption coefficients for a medium and air Dose warping ρm and ρair are the density of a medium and air, respectively (Radiotherapy) Dose warping is the process of deforming a refer- There are various types of dosemeters. ence dose distribution calculated on a reference image data set Related Articles: Ionisation chamber, Scintillation detector, (typically a planning CT scan) to a new image dataset (typically Semiconductor detector a daily CBCT image). This is done by warping the reference dose Further Readings: Delaney, C. F. G. and Finch, E. C. 1992. distribution using the same deformable image registration vector Radiation Detectors, Oxford University Press, Oxford, UK, field one would use to deform the reference image dataset to the pp. 333–338; Graham, D. T. and P. Cloke 2003. Principles of new image dataset. Radiological Physics, 4th edn., Elsevier Science Limited, New Whilst this method of dose warping can provide a quick esti- York, p. 328. mate of the new dose distribution for a patient on a particular treatment day, it is important to note that it is not a full recalcula- Dosimeter tion of the dose based on new anatomy so there will be inherent (Radiation Protection) The specific use of the terms dosimeter and errors in the displayed dose distribution. Areas in the dose distri- dosemeter in English refers to the following common meanings: bution with a large dose gradient or very large deformations will Dosimeter is usually a passive device used for patient or staff lead to greater uncertainty in the warped dose distribution. See dosimetry. This device accumulates the dose over a certain period in J. Radiol. Radiat. Ther. 2(2):1042 (2014) a schematic diagram of time and is retrospectively read. Example of dosimeters are for dose accumulation with deformable image registration (DIR). film badges, thermoluminescent dosimeters, etc. Each type of Related Article: Deformable image registration (DIR), Dose dosimeter requires specific calibration. accumulation Dosemeter is an active device which measures dose (or dose rate) in real time. Example of dosemeter is ionisation chamber Dose width product (DWP) with electrometer. (Diagnostic Radiology) The dose width product is a measure of Related Articles: Dosimetry, Film badge, Finger ring dosim- the radiation output, usually expressed as absorbed dose in air (or eter, Thermoluminescent dosimeter (TLD) Volume (%) 100 Volume (%) 100 Dosimetry 289 D osimetry systems Further Readings: Knoll, G. F. 2000. Radiation Detection and product. Normalised indices can be defined to allow comparison Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. with reference values to be made. 142–144; Saha, G. P. 2001. Physics and Radiobiology of Nuclear Further Readings: International Basic Safety Standards for Medicine, 2nd edn., Springer–Verlag, New York, pp. 71–72. Protection against Radiation and for the Safety of Radiation Sources. 1996. Safety Series No. 115, International Atomic Dosimetry Energy Agency, Vienna, Austria; 1990 Recommendations of (Radiation Protection) Dosimetry, in its meaning of measure- International Commission on Radiological Protection, Pergamon, ments of doses and related quantities is at the basis of radia- Oxford, UK, ICRP 60, 1991. tion protection as it is essential ‘to measure in order to protect’. Primarily it is essential to measure in order to use the appropri- Dosimetry protocol ate quantity/quality of radiation for diagnostic or therapy appli- (Radiotherapy) The dosimetry protocol (also known as a code of cations. Dosimetry covers a vast field including all aspects of practice) defines the procedures to be followed for the dosimetric personal and environmental dosimetry and monitoring for both calibration of clinical beams either in radiotherapy or in diagnostic ionising and non-ionising radiation; involving biological, physi- radiology. These protocols are written at the national (e.g. IPEM cal and biophysical aspects. It includes also external and internal [United Kingdom], AAPM [United States], DIN [Germany], etc.) D personal dosimetry for patients and staff, environmental and per- and international level (e.g. IAEA TRS 398, IAEA TRS 457). sonal monitoring (see also Occupational exposures), workplace These protocols are written also in association with primary stan- monitoring and accidental dosimetry. The definition of adequate dard laboratories which ensure high level of consistency through- measurement units is therefore essential in relation to the dif- out departments and among different countries. ferent kind of radiations. An appropriate dosimetry programme Abbreviations: AAPM = American Association of Physicists includes methodology of measurements and appropriate units in Medicine, DIN = Deutsches Institut für Normung, IAEA = based on the know-how of the kind of radiation beam, which is International Atomic Energy Agency and IPEM = Institute of going to be measured and its interactions. In this article, only as Physics and Engineering in Medicine. an example, the most usual radiation quantities and units needed Related Articles: Calculation of absorbed dose, Electron beam in the field of diagnostic radiology and corresponding dosimetry

dosimetry, Gray (Gy), Beam quality, Quality index, Reference are mentioned. depth, Dosimetry report Radiation Quantities and Units for X-Ray Diagnostic Dosimetry: Exposure is one fundamental concept. Its definition Dosimetry report evolved since its first use to measure a quantity of x-rays able (Radiotherapy) A dosimetry report can mean a number of differ- to produce a given electric charge in air (x-rays interacting with ent things. air produce ionisation). Modern definition is given according to It can be used to describe the results from an independent ICRU; SI unit is C/kg (former unit is the Roentgen). This quantity dosimetry audit of a radiotherapy department’s equipment. This is still used in several countries. Exposure rate is also defined. will relate the dosimetry values quoted by the department to those Absorbed dose deals with the complex process of energy measured by the external assessor using a different set of measur- transfer by ionising radiation to (or absorption by) matter. In a first ing instruments. stage kinetic energy is transferred from x-ray photons to charged Some people may use it to describe the documentation particles (concept of KERMA); in a second stage those charged received from a primary or secondary dosimetry standard labo- particles impart energy to matter (concept of absorbed dose). The ratory, although this is more commonly known as a calibration ICRU definition is given with corresponding SI units (Gray, for- certificate. merly rad). A relationship is given between exposure and absorbed A dosimetry report can also be used when an error in treat- dose (in air and for some photon energies and tissues). ment has been discovered to provide an assessment of the dosi- Mean absorbed dose applies to an organ considered with its metric impact, as a part of the protocol on accident. mass. Conversion factors are given between exposure and air or A further use of the term dosimetry report is in the area of per- tissue kerma, and between absorbed doses in air and soft tissue. sonal dosimetry, where the results from the person’s film badge Equivalent dose is helpful in quantifying the potentially harm- or TLD are documented over a set period of time (e.g. results for ful biological effects of the absorbed dose, modified by the radia- each month of a year). tion weighting factor, the SI unit is the sievert (formerly rem); it relates to the biological effect of radiation in an organ or tissue. The concept of effective dose allows expressing the detriment Dosimetry systems resulting from stochastic effects following exposure of different (Radiotherapy, Brachytherapy) tissues or organs. Effective dose is a weighted sum of equivalent Dosimetry Systems for Brachytherapy: Before the time of organ doses according to tissue specific weighting factors (WT computers, practical means of performing brachytherapy relied values are given), and it is useful to estimate the stochastic risk. on agreed rules based on clinical experience, pre-calculated dose The purpose of patient dosimetry is to assess doses (at the distributions, and a predictive dosimetry system. The coming of entrance or skin or in depth), usually with backscatter (present image guided brachytherapy has enabled true definition of vol- in the actual examination). Several concepts are used accord- umes of interest, as well as three-dimensional dose distribution ing to measuring technique: the incident dose (ICRP definition), calculations causing a decline in the use of classical methods entrance skin dose, backscatter factor (BSF), DAP (dose-area of dose prescribing. However it should be noted that there still product available on special devices). exists a wealth of clinical experience that has been gained using In the field of mammography dose is expressed in terms of these historical ‘systems’. Although image guided brachytherapy average glandular dose. is a huge step forward, care must therefore be exercised when In CT specific concepts are used to assess dose from an exami- changing from ‘classical’ to new image-based dose prescribing nation: multiple scan average dose, CT dose index, dose-length systems. Double contrast 290 Double focus tube Systems for Treatment of Cervix Carcinoma: In ICRU Report Further Readings: Gerbaulet, A., R. Pötter, J.-J. Mazeron and 38 ‘Dose and Volume Specification for Reporting Intracavitary E. van Limbergen, eds. 2002. The GEC ESTRO Handbook of Therapy in Gynaecology’ the following terminology is used: Brachytherapy, available at the ESTRO web site: www .estro .be; ICRU. 1985. Report 38. Dose and volume specification for report- ‘the term “system” denotes a set of rules taking into ing intracavitary therapy in gynaecology; ICRU. 1997. Report 58. account the source strengths, geometry and method of Dose and volume specification for reporting interstitial therapy, application in order to obtain suitable dose distributions Washington, DC. over the volume(s) to be treated. For reporting, the system includes recommendations for specifying the application Double contrast and possibly, as in the Manchester system, for calculating (Diagnostic Radiology) A double contrast study of the gastroin- the dose rate (or dose) at specific points.’ testinal tract uses a small amount of barium meal to coat the sur- faces and then air or gas to distend the colon for better visibility Three classical radium-based systems, all based on clinical expe- of potential lesions. D rience, are presented in short in this report: Double exposure radiograph • The Stockholm system, ‘intrauterine probe plus plate’ (Diagnostic Radiology) A radiograph that has been exposed a sec- • The Paris system, ‘intrauterine probe plus ovoids’ ond time by error. • The Manchester system, ‘intrauterine probe plus ovoids’ Examples of applications for these three systems are also given. Double focus tube Today (2009) radium is no longer used as a brachytherapy (Diagnostic Radiology) Usually an x-ray tube producing power source, and the manual radium loading techniques have been exposures would require large filament coil wire. However such replaced with remote controlled afterloading systems using 137Cs tube could not produce small effective focal spot, necessary for (low dose rate) and 192Ir sources (high dose rate and pulsed dose high resolution radiographs. In this case two wires are used (large rate). and small) allowing two effective focal spots. Usually rotational Many of the applicators used today for cervix cancer brachy- anode x-ray tubes use double focus. therapy are based on the classical systems, e.g. the ring applicator There are two main constructions of double focus x-ray tubes which is developed from the Stockholm system, see the article on – with the filaments one to the other and with the filaments one Intracavitary brachytherapy. above the other. In the first case both foci overlap, but in this Systems for Interstitial Brachytherapy: ICRU Report case the actual focal spots also overlap, causing overheating of 58 ‘Dose and Volume Specification for Reporting Interstitial some part of the anode surface and leading to shorter lifetime of Therapy’ states: ‘The term “system” denotes a set of rules which the x-ray tube (Figure D.72). In the second case both filaments takes into account the source types and strengths, geometry and have different actual focal spots, but the central beam is dis- method of application to obtain suitable dose distributions over placed which should be taken into consideration in centring the the volume(s) to be treated. The system also provides a means of radiographs). calculating and specifying dose. It is important to remember that while an implant may follow the source distribution rules of a system, it does not comply with the system unless the method of dose specification and prescription are also followed. In addition, if the implant rules are modified, the dose uniformity intended by the system may be compromised.’ Anode Three classical interstitial systems are mentioned: • The Manchester, Paterson–Parker, system (Ra) • The Quimby system (Ra) • The Paris system (192Ir wires, LDR) Cathode W Today (2009) image guided techniques are used more and more target for interstitial brachytherapy, e.g. the ultrasound guided interac- tive permanent seed implantations for prostate cancer brachy- therapy (see the articles on Interactive implant technique and interstitial brachytherapy). For an overview of dosimetry systems, both for intracavitary LF SF α α-anode and interstitial brachytherapy, the reader is referred to the GEC angle ESTRO Handbook of Brachytherapy, and references therein. Recommendation: Get to know the classical systems, especially the ones used in your clinic. Abbreviation: ICRU = International commission on radiation FF units and measurements. Related Articles: Intracavitary brachytherapy, Interactive BF implant technique, Interstitial brachytherapy, Paris system, Manchester system, ICRU reference point, Treatment planning FIGURE D.72 Double focus x-ray tube with overlapping filaments (one systems – Brachytherapy to the other). Double scattering 291 Dragonfly CT Note on Figure D.72 that projections of both cathode fila- lesions responsible for the induction of genetic changes in ment coil wires (for large/broad focus BF and for small/fine focus mammalian cells, including chromosomal abnormalities and FF) overlap over part of the actual focal spot, which appears gene mutations. overloaded. Cells possess a complex set of signalling pathways for rec- Related Articles: Stationary anode, Rotating anode, Target, ognising DNA damage and initiating its repair. The ATM gene Line focus principle, Biangular anode disk, Focal spot actual, is one of the main sensors of DNA damage, which activates by Focal spot effective, Focal spot phosphorylation a variety of proteins involved in cell cycle control and DNA repair. Unrepaired or misrepaired DSB lead primarily Double scattering to large-scale genetic changes that are frequently manifested by (Radiotherapy) In passive scattering, lateral spreading of the chromosomal aberrations. proton beam is achieved by placing two scatterers in the beam. Further Readings: Hall, E. 1994. Radiobiology for the The first (upstream) scatterer is a set of flat foils. The second Radiologist, 4th edn., J. P. Lippincott Company; Khan, F. 1994. (downstream) scatterer is approximately Gaussian in profile: The Physics of Radiation Therapy, 2nd edn., Williams and protons near the centre must be scattered more than protons fur- Wilkins; Nias, A. 2000. An Introduction to Radiobiology, 2nd ther out in order to flatten the field. High atomic number mate- edn., John Wiley & Sons. D rials scatter the beam with minimum energy loss, while low atomic number materials decrease beam energy with minimum Downscatter scattering. Therefore scattering and beam energy can be con- (Nuclear Medicine) Downscatter refers to an effect that is com- trolled through a combination of low and high atomic number mon when acquiring simultaneous transmission and emission scatterers. images. The transmission image is used for attenuation correction Related Article: Passive beam scattering and emission images are acquired using a different radionuclide, Further Reading: Khan, F. M. and J. P. Gibbons. 2014. Khan’s hence a different emission energy. The two scans can therefore the Physics of Radiation Therapy, 5th edn., Wolters Kluwer be acquired by counting events in two different energy windows. Health. Inevitably a number of photons from the high energy emitting radionuclide undergo Compton scattering in the object before Double-strand break being registered by the detector. The energy registered in such (Radiotherapy) Directly opposite breaks (or even more distantly an event is lower than the original emission energy and might be opposite breaks) in the DNA double helix can result in a dou- falsely registered in the lower energy window. Another cause of ble-strand break (DSB) of DNA, as it is cleaved into two seg- downscatter is poor crystal energy resolution and/or small differ- ments (Figure D.73) as a result of a harmful event, e.g. exposure ence in emission and transmission photon energy, such that events to ionising radiation. In other words, DNA double-strand break in the two windows will be ‘misplaced’ and they are registered in means that both strands of the DNA helix are severed at a similar the ‘wrong’ window. location. Downscatter can also be a problem when acquiring the scans DSB Damage: When a mammalian cell nucleus is exposed sequentially because the radiopharmaceutical is sometimes to 1 Gy (= 1 J/Kg) of low LET (Linear Energy Transfer: the rate administered prior to the examination (sometimes several hours of the energy loss of a radiation) radiation the damage sustained in advance) and activity is therefore present during the transmis- consists of approximately 1,000 DNA single-strand breaks (SSB) sion scan as well. Downscatter can be estimated and corrected for and 40 double-strand breaks (1 DSB per chromosome per Gy). in the same fashion as scatter correction (see separate article). An Figure D.73 shows examples of SSB (diagrams (B) and (C)) and energy window between the transmission and emission windows DSB (diagram (D)) that may occur to

a DNA. A DSB results from can be used to estimate the level of downscatter. a combination of two SSBs to a single base pair or to two base Related Articles: Positron emission tomography (PET), pairs located close to each other. As is demonstrated in (D), a DSB Scatter correction in SPECT, Single photon emission computed leads to DNA breakage. tomography (SPECT) A double-strand break is believed to be the most important Further Reading: Cherry, S. R., J. A. Sorenson and M. E. radiation-produced lesion in chromosomes since the interaction Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, of two double-strand breaks may result in cell killing, mutation Philadelphia, PA, p. 313. or carcinogenesis. Consequence of DNA Damage: Although ionising radia- DQE tion can produce a broad spectrum of DNA lesions including (Diagnostic Radiology) See Detective quantum efficiency (DQE) damage to nucleotide bases, unrepaired or misrepaired DNA double-strand breaks (DSB) are thought to be the principal Dragonfly CT (Diagnostic Radiology) Industrial CT is mainly used in research departments and quality control labs due to the time required to perform an inspection scan of a certain manufactured part. Using robot-based automation for part loading and fast sample rotations for scans, industrial inline CT inspections are now available on the production line. Dragonfly CT is a macro-CT technique in which the detector and x-ray source have fixed positions and a robotic arm is used to load and rotate samples for image acquisi- tion (see Figure D.74). Scan times with this methodology have FIGURE D.73 Schematic diagram of normal DNA segment, DNA sin- reached 0.8 s. gle-stand breaks (SSB) and DNA double-strand break. Related Articles: Macro-CT, Industrial CT Drift 292 DRL (Diagnostic reference levels) Drift mobility (Radiation Protection) The external electric field applied to the radiation detector (e.g. gas-filled chamber or semiconductor) pro- duces a drift. The free charge carriers (electrons, positive and negative ions in gas-filled radiation detectors, electrons and holes in semiconductor detectors) are accelerated in the direction deter- mined by the electric field. The drift mobility u for ions in gas-filled detectors is equal to v × p u = [m2atm/(Vs)] E where v [m/s] is the drift velocity of ions D p [atm] is the gas pressure E [V/m] is the electric field strength The values for mobility of positive and negative ions are between 1 and 1.5 × 10−4 m2atm/(V s) in the a gas at 1 atm pres- sure and with an applied electric field E about 104 V/m. The FIGURE D.74 Schematic diagram of the source, sample and detector mobility of free electrons in the same gas is about 103 greater setup for a Dragonfly CT system. because of their much lower mass. The drift mobility for electrons un and holes up is defined as v Drift u n n = [m2 /(Vs)] (Radiation Protection) The movement of charge carriers pro- E duced by ionising radiation (electrons and holes in semiconductor devices, ions and electrons in gas-filled radiation detectors e.g. v u p p = [m2 /(Vs)] ionisation chamber) under influence of the external electric field E E [V/m] created between the electrodes is called drift. The drift where current density J [A/m2] is equal to the sum of the negative charge vn and vp [m/s] is the velocity of electrons and holes, respectively carriers (electrons, negative ions) Jn and positive carriers (holes, E [V/m] is the electric field strength positive ions) Jp: The following table demonstrates how the drift mobility of J = Jn + J p electrons and holes in silicon and germanium is affected by the temperature of the semiconductor: where Jn = nnqnvn and Jn = npqpvp nn, np are the number of negative and positive carriers per cm3, Silicon (cm2/(V s)) Germanium (cm2/(V s)) respectively qn and qp are the electric charge of the carriers Temperature (K): Electron Hole Electron Hole vn and vp are their respective velocities 300 1,350 480 3,900 1,900 Introducing the mobility u [m2/(V s)] of a charge carrier as a 77 21,000 11,000 36,000 42,000 ratio of its velocity v and electric field strength E, the dependence between current density and electric field E can be expressed as As shown in the table, the drift mobility of electrons and holes in J = (nnqnun + npqpup )E both silicon and germanium is significantly increased by reducing the temperature of the material; from ambient room temperature The net velocity of the charge carriers is a combination of the (∼300 K) down to that of liquid nitrogen (77 K), improving the thermal velocity (i.e. about 107 cm/s for silicon at room tem- performance of such materials when used as radiation or imaging perature) and the drift velocity produced by the applied elec- detectors. tric field E. The mobility of charge carriers remains constant Related Articles: Drift, Gas-filled radiation detectors, and small compared to the thermal velocity over wide ranges Semiconductor detector of electric field strengths and is solely dependent on the kind Further Readings: Knoll, G. F. 2000. Radiation Detection of detector. and Measurement, 3rd edn., John Wiley & Sons, Inc., New Related Articles: Gas-filled radiation detectors, Semiconductor York, pp. 133, 356–357; Lutz, G. 2007. Semiconductor Radiation detector Detectors. Device Physics, 2nd printing, Springer Verlag, Berlin, Further Readings: Knoll, G. F. 2000. Radiation Detection Germany, pp. 16–17. and Measurement, 3rd edn., John Wiley & Sons, Inc., New York, p.133; Sze, S. M. 1985. Semiconductor Devices. Physics and DRL (Diagnostic reference levels) Technology, John Wiley & Sons, New York, pp. 35, 62. (Radiation Protection) See Diagnostic reference levels (DRL) Dropping load 293 Dual energy digital radiography Dropping load from fluid with high signal intensity and bone with lower signal (Diagnostic Radiology) See Falling-load generator intensity. Related Articles: Fast imaging with steady state (FISP), PSIF Drying in film processing (FISP reversed), SSFP, Steady state free precession, T2-weighted (Diagnostic Radiology) The final step performed by a film pro- Further Reading: Hardy, P. A., D. Thomasson, M. P. Recht cessor. The film passes through a chamber of circulating warm air and D. Piraino.1996. Optimization of dual echo in the steady state before dropping out of the processor. (DESS) free-precession sequence for imaging cartilage. J. Magn. Reson. Imaging 6:329–335. Dry laser imager (Diagnostic Radiology) See Laser film printer Dual energy CT (Diagnostic Radiology) Conventional single-energy CT acquisi- DSA (digital subtraction angiography) tions provide limited information on material composition of tis- (Diagnostic Radiology) See Digital subtraction angiography sues having similar attenuation coefficients and hence a similar (DSA) Hounsfield number, even if there is a difference in mass-attenua- tion coefficients and elemental composition. D Dual energy CT (DECT) has been developed over the years DSI (diffusion spectrum imaging) to provide information about tissue composition. This tech- (Magnetic Resonance) See Diffusion spectrum imaging (DSI) nique has been implemented by CT scanner manufacturers in a number of ways; dual source CT scanners, single source scan- DTI (diffusion tensor imaging) ners and detector-based spectral CT. The tube source energies (Magnetic Resonance) See Diffusion tensor imaging (DTI) implemented by each manufacturer may vary but indicative val- ues include 80kVp and 140kVp for the low and high energies DTPA respectively. (Nuclear Medicine) See TC-99M-DTPA Dual source DECT systems use two x-ray tube (one high energy and one low energy) and corresponding detectors to Dual echo steady state (DESS) acquire two sets of images; single source DECT systems use rapid (Magnetic Resonance) The dual echo steady state (DESS) pulse kilovoltage switching to acquire, almost simultaneously, two sets sequence belongs to the family of steady-state sequences, and its of images. Detector-based spectral CT systems use two layers main advantage is the differentiation between fluid and cartilage. within the detector bank; the superficial layer absorbs the low The basic idea of the DESS pulse sequence is to collect both energy x-ray photons and the deep layer absorbs the high-energy the SSFP-FID and the SSFP-Echo signal during two separate photons. Figure D.76 illustrates the main types of DECT. readout periods after each RF-pulse and then to combine the Single source rapid kilovoltage and detector-based spectral result into one final image, i.e. to combine the FISP and the PSIF DECT systems use the two-material decomposition analysis in pulse sequence into one. The benefits from this pulse sequence the projection domain to create material-specific images from the is the gain in signal-to-noise ratio, coming from sampling two two datasets before tomography images are reconstructed. This echoes and the heavily T2-weighted contributions from the SSFP- algorithm assumes that the volume is made of only two prese- echo signal which significantly increases the signal for fluid. The lected materials. The images produced provide concentration DESS pulse sequence is advantageous for joint imaging where it information. In clinical practice water and iodine are the pair provides high signal intensity for fluid and good delineation of used typically. cartilage. In Figure D.75, it is demonstrated how the DESS pulse Dual source DECT systems use the three-material decom- sequence delineates cartilage with intermediate signal intensity position analysis in the image domain to generate material- specific images from the two data sets. As there are only two image data sets to use three material decomposition is made by using the constraint of mass conservation, i.e. the mass within a voxel is equal to the sum of the individual mass fractions of the three materials. At first the two materials technique is used to compute the mass fraction of two materials and then the missing fraction is attributed to the third material. This algorithm provides information such as attenuation and iodine concentration. Related Article: Dual energy index Dual energy digital radiography (Diagnostic Radiology) Dual energy digital radiography is a new technique used to improve the visibility of details primar- ily in chest imaging although it can also be employed in other examinations. Technically, two images are acquired at different x-ray ener- gies. Soft tissue and bone have different absorption properties depending on the beam spectrum. This aspect is utilised to pro- duce three separate images: a conventional planar image and, FIGURE D.75 Cartilage delineation obtained with the DESS pulse derived from this, an image with only soft tissue and an image sequence. with only bone and calcified tissue. Dual energy imaging 294 Dual energy subtraction D FIGURE D.76 Diagram indicating the main three types of dual energy CT (DECT) systems; dual energy source DECT (left), energy switching DECT (centre), two-layer detector DECT (right). Two different technical solutions exist to discriminate soft tis- sue and bone, depending on the digital detector used. TABLE D.5 DEI Values for a Number of Tissue Types • Using photostimulable phosphor plates, a single expo- for Dual Energy CT Acquisitions sure is taken using a dual detector made by two inte- grated CR plates, separated by a thin copper filter. This Performed at 80 kVp and 140 kVp technique avoids the risk of blurring due to patient Tissue DEI movement, but the back plate receives a lower dose than Bone 0.1148 the front plate and this increases the noise in this image. Lung –0.0021 • Using a direct digital detector two exposures are taken Fat –0.0194 in rapid succession, selecting different kV values. This solution is possible only on equipment able to acquire two images in a very short temporal interval. It offers the advantage to select very different x-ray Dual energy index energies (typically the consecutive kV selections are (Diagnostic Radiology) In dual energy CT (DECT) the dual 60 and 120 kVp) and to produce images with a low energy index (DEI) has been introduced to help differentiate noise level. The disadvantage is the risk of voluntary between materials. It is defined as and especially involuntary patient movement between the exposures. HUkVpL - HU DEI = kVpH Dual energy imaging HUkVpL + HU kVpH + 2000 (Diagnostic Radiology) In CT scanning dual energy (dual-energy) imaging refers to the acquisition of images at two different x-ray where HUkVpL and HUkVpH are the CT Hounsfield units of the mate- energies. This enables improved differentiation and characterisa- rial of interest, at a low and high kVp respectively (e.g. 80kVp and tion of materials in the imaged object. Conventional CT scanning 140kVp). DEI is affected by the material’s density and Zeff. This is performed with a single energy spectrum of x-rays. The infor- aids in the identification of chemical composition of the region of mation obtained is of the

distribution of linear attenuation coef- interest and estimation of the effective atomic mass numbers. It ficients and does not uniquely identify the materials. allows the identification of pure materials up to an atomic number Dual-energy CT is usually achieved by measuring the attenua- of 55. Clinical uses include differentiation between kidney stones tion using two different energy spectra, e.g. one at 80 kVp and the with and without uric acid composition. Examples of DEI values other at 140 kVp, but can also be performed with one energy spec- for a number of tissue types are given in Table D.5. Water (or trum and a dual-layered detector. For dual x-ray spectra acquisi- water like tissue) would have a DEI of 0. tion, one approach is the use of a dual source system, with the Abbreviation: Zeff = Effective nuclear energy. two x-ray tubes operating at different kilovoltage potentials (kVp). Related Articles: Dual energy CT, Hounsfield units Another method, with a single source system, is to switch the tube Further Reading: Patino et al. 2016. Material separa- kVp between alternate projections within a rotation. The two data tion using dual-energy CT: Current and emerging applications. sets can then be analysed for differences in attenuation enabling a Radiographics 36 (4). better characterisation of selected materials. One clinical application of dual-energy CT is better differen- Dual energy subtraction tiation between bone and iodine contrast material, allowing for (Diagnostic Radiology) Dual energy subtraction is an x-ray imag- improved bone removal in CT angiography images. Some other ing method that can be used to separate tissues, especially soft applications include diagnosis of kidney stone type, and discrimi- tissues and bone. Because of its calcium content, bone has a rela- nation between calcified plaque and contrast material in coronary tively high x-ray attenuation at the lower photon energies. The angiography. difference between bone and soft tissue attenuation is less at the Related Articles: Computed tomography, Dual source CT higher photon energies. Dual energy x-ray absorptiometry 295 Dual filament tube FIGURE D.77 A conventional radiograph (left) compared to a bone-selective image produced with a dual energy subtraction method. A typical procedure would produce 2 images with 2 kV values, such as 60 and 120 kV. D An image subtraction can then be used to produce both bone- selective and soft tissue-selective images in addition to conven- tional radiographs (Figure D.77). Dual energy x-ray absorptiometry (Diagnostic Radiology) Dual energy x-ray absorptiometry (DEXA or DXA) (Figure D.78) is generally used to obtain the mass of one material in the presence of another. This is done through the prior knowledge of their x-ray attenuation at different energies. In clinical practice, DEXA’s main application is body mineral density measurements for the assessment of fracture risks and diagnosis of osteoporosis. DEXA involves acquiring two images using a low and a high x-ray energy beam and spatial registration of the two attenuation maps. The tube source energies implemented by each manufac- turer may vary but indicative values include 100 kVp and 140 kVp for the low and high energies respectively. The principle can only be used to solve for two materials simultaneously. However, three FIGURE D.78 Dual energy x-ray absorptiometry scanner. Note that due assumptions can be made to quantify three materials (bone, lean to the very low scatter radiation dose rate and depending on the room size no radiation shielding may be required for the operator. mass and fat mass): • The exponential attenuation process describes x-ray transmission through the body for the two energies. where Rs is the mass attenuation coefficient ratio value for soft • Pixels in the images produced can represent two com- tissue (at the two energies) measured for tissue surrounding but ponents, i.e. either soft tissue and bone mineral or lean not containing the bone. Calibrations should be performed with mass and fat mass. specialised phantoms routinely for each particular examination • X-ray properties of the tissue near the bone structure and DEXA system. can be used to predict composition and x-ray properties T and Z scores are routinely used for the diagnosis of osteopo- of the tissue overlying the bone. rosis and low bone mass in young adults and children respectively. The reference values used to calculate these scores are derived from databases of local populations. The attenuation equation for each energy would result in: Dual filament tube é E E ù -êæ m ö æ ç ÷ x m ö s +ç ÷ xb ú ê (Diagnostic Radiology) Each classical x-ray tube with two effec- IE = I ëè r øS è r ø ú 0e b û tive focal spots requires two filament wires – a large filament wire (producing broad focus) and a small filament wire (producing where the E subscript represents either the high or low energy fine focus). Usually these dual filament tubes are with rotational x-ray beams and the s and b subscripts represent soft tissue and anode (including all biangular anode tubes). See Figure D.72 and bone, μ/ρ is the mass attenuation coefficient of the material and ξ Figures R.60 through R.63. is the areal density. Solving the above equation for the areal den- These two wires can sit in one focussing cup, or in different sity of bone (aBMD) leads to: focusing cups (Wehnelt electrode) – see Figure C.16. The two wires can be positioned one to the other or one above the other. In the first case the two filaments create overlapping of actual æ 1 ö æ 1 R In H In L ö ç focal spots. This could lead to cracks (due to thermal stress), and s ç è 1H ÷ - L ÷ x = aBM D = 0 ø è 10 ø will decrease the life of the tube. When the two filaments are b L H æ m ö æ m ö one above the other, this thermal problem is avoided, but the two ç ÷ - ç R r ÷ s effective focal spots are slightly displaced, which should be cor- è r øb è øb rected at beam centring. Dual foil system 296 Duplex ultrasound Related Articles: Cathode, Filament, Rotating anode, Double Dummy sources focus tube, Target, Biangular anode disk, Focal spot actual, Focal (Radiotherapy, Brachytherapy) Most remote controlled afterload- spot effective ing devices for HDR or PDR brachytherapy are equipped with a miniature high strength source attached to a flexible cable which Dual foil system allows source deployment and a stepping movement. (Radiotherapy) See Electron dual scattering foils As part of the safety system, these afterloaders are equipped with a second cable and non-radioactive metal pellet whose Dual source CT mechanical characteristics are identical to the active source cable (Diagnostic Radiology) The dual source CT (DSCT) system has and pellet. This is called a dummy (or check) source. This dummy two sets of tube-detector pairs mounted at an angular offset of 95° source is deployed ahead of the active one to check that the active on the rotating gantry. DSCT is schematically shown in Figure D.79. source will be able to negotiate the catheters or other source carri- One detector (detector A) covers a 50-cm scan field of view ers and guides. The treatment is not allowed to start, if the dummy (SFOV). To preserve compact system geometry, the second detec- source is unable to do this. tor (detector B) covers a smaller SFOV – 26–35 cm (it depends Abbreviations: HDR = high dose rate and PDR = pulsed dose D on the generation of dual source CT). Both tubes can be operated rate. independently with respect to voltage and current settings. Related Article: Remote after loading The first key benefit of DSCT is improved temporal resolu- tion. This is specifically useful for cardiac scanning, since each Duplex ultrasound measurement system needs to contribute only 90° of data in a (Ultrasound) Duplex ultrasound refers to an ultrasound device half scan or 45° of data in a quarter scan, resulting in an effective which combines a B-mode image with pulsed wave (or sometimes temporal resolution of 83 or 165 ms, established for a gantry rota- continuous wave) Doppler ultrasound. The position of Doppler tion time of 0.33 s. beam can be displayed on the B-mode image, enabling position- The second key benefit of DSCT is that it offers high tube ing of the beam and, for pulsed Doppler, the sample volume to power for obese patients. This is achieved by adding data from investigate a particular area of tissue. If the direction of the ves- detectors A and B in the image reconstruction unit. The high tube sel under investigation is clear, duplex ultrasound can enable the power also enables long-scan range, e.g., whole-body CT angio- operator to determine the beam/vessel angle, thereby permitting graphic studies at sub-millimetre resolution. measurements of velocity from the Doppler spectrum. The third key benefit of DSCT is that it can perform dual In a duplex scanner the B-mode image may be frozen while energy CT, scanning with low and high kV combinations. Doppler investigation is undertaken, or the image may be updated Second and third generation DSCT employs a tin filter (selective at a user-selected interval. The B-mode image may run concur- photon shield, as shown in Figure D.79) for increased x-ray energy rently with the Doppler but the B-mode frame rate and the tem- spectra separation. After acquisition, the data delivered by detector poral resolution of the Doppler sonogram may suffer due to the A and detector B is reconstructed separately, and two different CT limited time for each. numbers are assigned. By using DSCT for dual energy CT scanning, The term triplex scanning refers to concurrent B-mode, colour misregistration between low and high tube voltage images (due to flow and spectral Doppler in a scanner. Triplex scanning has the patient motion between two consecutive scans) can be avoided. advantage that movement of vessel can be tracked. However, the However, a drawback of DSCT system is additional x-ray scat- performance of each mode may be compromised (e.g. by limiting ter originating from another set of a tube-detector pair. Therefore, colour flow update and size of colour flow ‘box’), due to the lim- scatter correction technique needs to be applied when recon- ited amount of time available for each. structing DSCT images. Duplex scanning first came into use in the mid to late 1970s. Related Article: Dual energy CT The term duplex is still used colloquially even in colour flow imaging and describes a scan in which B-mode and Doppler imaging is used (Figure D.80). FIGURE D.80 Duplex ultrasound of an internal carotid artery (ICA) showing the location of the sample volume for pulsed wave spectral FIGURE D.79 Dual source CT system. Doppler in the centre of the proximal ICA and the resulting sonogram. Duplicating film 297 Dynamic imaging Duplicating film movement and several treatment channels. The source movements (Diagnostic Radiology) Duplicating of x-ray radiographic film is are computer controlled, both the source step size between stop usually made with a special duplicator – a light source where the positions and the dwell time at each stop position. original x-ray film is placed together with a duplicating film (usu- Figure D.81 shows two dose distributions calculated with ally with single emulsion). The light (often ultraviolet) from the seven stop positions, step size 1.0 cm. In Figure D.81a, equal duplicator passes through the original film (what modulates it) dwell times have been used to give a dose of 10 Gy at a distance and after this exposes the duplicate film. This way an exact (and of 2 cm from the central stop position. In Figure D.81b, the dwell negative) copy of the original is produced. times have been optimised to give 10 Gy at a distance of 2 cm Further Reading: Thompson, M., M. Hattaway, D. Hall and from all stop positions. Notice the tapering shape in Figure D.81a S. Dowd. 1994. Principles of Imaging Science and Protection, and the prolonged dwell times at the first and last stop positions W.B. Saunders Company, Orlando, FL. in Figure D.81b. Related Article: Remote afterloading unit Duty cycle (Magnetic Resonance) The duty cycle is basically a measure of Dynamic aperture how heavily the system can be loaded with respect to power (heat- (Ultrasound) In a linear array system, the receiver aperture (num- D ing) over time. ber

of active elements) is expanded with increasing investigation In a general way duty cycle can be defined as the proportion of depth in a process referred to as dynamic aperture. The reason for a specified time period (Ton + Toff) during which a device or system using this is that the lateral resolution is directly proportional to is operated (Ton): the ratio of imaging depth to array aperture (commonly referred to as the F-number). To maintain the lateral resolution, more æ T ö and more elements are included in the aperture as the distance Dutycycle = 100 × ç on ç ÷÷(%) increases, at least up to the point where the number of avail- è Ton + Toff ø able channels in the hardware is the limiting factor. The term is The way of defining the duty cycle may differ between vendors used in receive mode, but correspondingly the aperture can also but typically it is reported to be 100% for an MRI system. A be increased in transmit, but only in a discrete number of steps, more specific description of the duty cycle than the definition resulting in a number of focal zones. This process is however not used here would have to reflect its dependence on how the sys- dynamic as it is for receive, when the aperture opens up for deeper tem dissipates power over time in relation to how the system is echoes from the same imaging line. operated. In practice it might therefore simply be a good idea On Figure D.82 the echoes from a shallow reflector A return to to verify the reported duty cycle by finding out if the MRI sys- the transducer with different time/phase changes between trans- tem can execute any pulse sequence with any imaging param- ducer elements when compared with those from a deeper reflector eters for as long a time as desired without interruptions for heat B. The delay and sum instrumentation uses a limited aperture for evacuation. echoes from A. Echoes from B arrive slightly later and the sys- For MRI equipment the radiofrequency (RF) and gradient tem uses an increased number of elements with different delays to systems are the ones that essentially dissipate heat and therefore achieve effective focussing. needs to be cooled. A high tolerance to heating of the system is of course favourable. Water cooled gradient systems can, e.g. pro- Dynamic focusing vide a higher duty cycle than air cooled systems, which is nor- (Ultrasound) Dynamic focusing is a technique applied in receive mally required with modern MRI scanners that often utilise fast mode to enable a receive focus over the entire length of the inter- imaging techniques. rogated B-mode-line. In Figure D.83, an echo is produced by a scattering object in the interrogated direction. To focus at that Duty cycle depth, delays are introduced to the signals originating from the (Ultrasound) The duty cycle describes the fraction of time for individual transducer elements. The delay corresponds to the which a transducer, operating in pulsed mode, transmits ultra- propagation delay, and makes the signals appear to have been sound. It is the ratio of the pulse length to the time between received simultaneously. After summation, these signals add con- pulses. Duty cycle is also referred to as Duty factor. structively, and produce a large signal. Objects out of focus add destructively and are thus suppressed. In dynamic focusing, these delays are adjusted for each point in the interrogated direction so EXAMPLE: that an optimal focus along the entire line is achieved. On Figure D.83 as the wavefront from a scattering object A physiotherapy ultrasound machine transmits pulses approaches the transducer, the wavefront is curved. By applying of 2 ms duration followed by 8 ms pauses. The duty a delay to the individual signals that corresponds to the expected cycle is 1:4. This may also be described as a duty propagation delay at a given distance, signals arriving from that cycle of 20%. distance are summed to produce a large echo. Dynamic imaging In diagnostic ultrasound applications the duty cycle is typi- (Nuclear Medicine) Dynamic imaging refers to a set of images cally less than 1%. acquired in a region of interest over a period of time. For exam- Related Article: Pulsed mode ple, dynamic image sets can be used to study the bio-kinetics of tumour targeting agents. By studying the amount of activity in Dwell time an organ or organ region it is possible to draw conclusions about (Radiotherapy, Brachytherapy) A typical remote afterloading the uptake, redistribution and clearance of targeting agent in the unit for high dose rate and pulsed dose rate brachytherapy has one selected region. The bio-kinetics can then be modelled by a time- single small source of high specific source strength with stepping activity curve used in dosimetry calculations. Dynamic jaw collimation 298 Dynamic radiosurgery D (a) (b) FIGURE D.81 (a) Seven stops, equal dwell times, 10 Gy at 2 cm distance from central stop position. (b) Seven stops, optimised dwell times, 10 Gy at 2 cm distance from all stop positions. Dynamic jaw collimation Dynamic multileaf collimation (Radiotherapy) Dynamic collimation is achieved when the beam (Radiotherapy) See Multileaf collimation collimation is changed during the time the beam is delivering treatment. In this case the beam aperture size changes either by Dynamic radiography movement of a solid jaw or by using the multi-leaf collimator (Diagnostic Radiology) Dynamic radiography is another name for (MLC). The latter is sometimes referred to as dynamic multi-leaf x-ray fluoroscopy (a method for visualising the organs in motion). collimation (DMLC). Related Article: Fluoroscopy Related Articles: Collimation, Dynamic wedges, Dynamic multileaf collimation, Asymmetric jaws Dynamic radiosurgery Further Reading: Walter, J. and H. Miller. 2003. Textbook (Radiotherapy) Dynamic radiosurgery is a special technique of Radiotherapy Radiation Physics, Therapy and Oncology, 6th where the gantry moves during the treatment with the beam on. edn., Churchill Livingstone, London, UK. There are two approaches: Dynamic range 299 Dynamic range quantifying the dynamic range is usually the deciBel (dB), a unit of dimension given in decadic logarithms. The lower bound for converting the energy imparted to the image detector to a signal is usually defined by detector characteristics such as the DQE. Delay and sum Delay and sum Another quantity that defines the lower bound of the dynamic range is bias noise, which is also connected to the signal-to-noise ratio (SNR) of a signal. Histograms: A measure for optimal use of the dynamic range in a medical image is the histogram, which represents the dis- tribution of the measured signal over the dynamic range in an image. An optimum histogram shows a smooth and even distri- bution with little signal in the low and high end of the dynamic range. If the dynamic range is not fully utilised, or if large parts of the histogram cumulate at the lower or upper bound of the dynamic range, it is likely that the signal resolution is either not D fully utilised or that usable signals are not recorded due to satura- A tion or detector inefficacy. Transfer Functions: In general image processing, the trans- fer function of an input signal Iin to an output signal Iout, which gives a graphic illustration of the dynamic range in terms of a response curve, is governed by the gamma-value. The gamma- value gives a measure of the slope the transfer function and is B used in film-based radiography as a measure of the contrast behaviour of the film used. A high-contrast film usually features a steep gradient; therefore, small input contrast detail differences FIGURE D.82 Dynamic receive aperture. are recorded over a large span of the dynamic range available. As a consequence, such media is easily subject to over- or under- exposure. In medical imaging, the most important transfer func- tion operation is windowing of images, where a certain part of the histogram known to contain anatomical structures of interest such as soft tissue or the bony skeleton is selected and displayed within a smaller dynamic range; by applying a windowing opera- tion, small contrast detail can be visualised in a manner that is optimal to human perception. Modelling the Transfer Function: A simple mathematical model for a symmetric transfer function can be derived from the logistic function. It is a sigmoid type of mathematical function given as FIGURE D.83 Dynamic focusing. D *1 I(r) = (1+ e-(r-c)/d ) Linac-Based Dynamic Radiosurgery: This is radiosurgery on a linac using the arc technique. Often several arcs may be used I(r) is the resulting signal and couch movement may be used. r is the input signal Four-Dimensional Dynamic Radiosurgery: This is the use D is the numerical range for recording the signal of radiosurgery techniques such as the Cyberknife robotic linac c is the window centre (the location of the turning point of the to track target motion in radiosurgery. This is particularly use- sigmoid) ful for treatment sites which exhibit intra-fraction motion, such d is the width of the window as the lung. Related Articles: Stereotactic radiosurgery, Gamma knife, Dynamic range Robotic linac, Arc therapy, Stereotactic frame (Ultrasound) In general, the dynamic range is the ratio between the smallest and largest possible values of a quantity. The Dynamic range dynamic range is usually quoted in decibels, i.e. 10 log(P1/P0) (General) In signal processing, the term dynamic range usually for power related measures, and 20 log(A1/A0) for amplitude refers to the ratio of the minimum and maximum of a measurable related measures. For instance the dynamic range of a 10 bit quantity. In medical imaging, these optima are usually the mini- A/D-converter can be found as 20 log(210/1) which approxi- mum energy necessary to cause a measurable signal in a detector, mately is 60 dB. Ultrasound scanners often quote the dynamic and the maximum energy that can be measured without saturation range on-screen which includes effects of multiple elements, of the signal. In digital imaging devices, this saturation can also time gain compensation (TGC) and non-linear amplification. be limited by the chosen image depth (i.e. the maximum analogue For instance an aperture using 100 elements, where each chan- value that can be converted to a given data type). The unit used for nel is using a 10 bit A/D-converter can achieve a dynamic range Delay Summation Dynamic range 300 Dynamic range of 40 + 60 = 100 dB (i.e. the minimum detected signal would signal response) can be produced in an acquired image. Outside be a 1-bit change on one channel, whereas the maximum is the the dynamic range, areas of differing photon attenuation within maximum change on all channels). TGC-compression and non- an imaged object (and hence differences in the number of photons linear amplification (log amplification) can increase this num- reaching the detector) will not be presented as different greyscale ber further. levels. If exposure is too low then the image will be dominated by For ultrasound scanner operators, the term dynamic range is noise, if exposure is too high, the detector will be saturated across used to describe the on-screen range of echoes displayed. Many its entire face. scanners permit the operator to alter the displayed dynamic range, The range of pixel values a detector is able to represent is e.g. to have a limited dynamic range of echoes, increasing con- termed its bit depth. For example, if a detector has a bit depth of trast and suppressing weak echoes or alternatively increasing the 8, it can produce 2 to the power 8 (28 = 256) shades of grey from range to highlight weak echoes. The effect of changing this is white to black. The dynamic range of a detector can be described shown in Figures D.84 and D.85. as the ratio between the exposure that gives the highest pixel With a reduced dynamic range 40 dB (Figure D.84), there is value (i.e. saturation, 256 in this case) and the exposure that gives greater contrast within the kidney. However, echoes from the cor- the lowest pixel value above the noise threshold of the detector D tex are weak. In the lower image (Figure D.85), the weak echoes (the noise of the detector with no exposure). from the cortex are now displayed as dynamic range is increased Digital detectors have a wide, linear dynamic range, mean- to 65 dB. ing a wide range of

exposures will produce an image, whereas screen film systems with a narrow and S-shaped dynamic range Dynamic range (Figure D.86). Although this linear dynamic range may seem (Diagnostic Radiology) The dynamic range of a detector is the solely advantageous, namely, there is less restriction on the dose range of incident exposures where contrast (i.e. differences in to the detector required to produce an image, this may lead to exposures (and hence patient doses) that are far higher than nec- essary to attain suitable image quality. Previously, with screen film systems, over-exposures would result in a saturated film and would be easily identifiable. Related Articles: Bit depth, Dynamic range (Magnetic Resonance), Dynamic range (Ultrasound) Further Readings: Körner, M. et al., 2007. Advances in digital radiography: Physical principles and system overview. RadioGraphics 27:675–686; Samei, E. 2003. Performance of dig- ital radiographic detectors: Quantification and assessment meth- ods, advances in digital radiography. RSNA Categorical Course in Diagnostic Radiology Physics, pp. 37–47. FIGURE D.84 Reduced dynamic range to 40 dB. FIGURE D.86 The signal response of a digital detector and screen film system against incident dose, displaying the far wider dynamic range of FIGURE D.85 Increased dynamic range to 65 dB. digital systems. Dynamic receive focusing 301 Dynamic susceptibility contrast MRI Dynamic receive focusing for a very short time (on the order of 15 ms per scan). This system (Ultrasound) An ultrasound beam can be focused during either can scan the heart, but was very expensive and difficult to build, signal transmission or receiving, or both. As the system transi- so only prototypes were produced. Later another ultra fast CT tions between transmission and receiving the proportion of crystal system was made (electron beam CT, e.g. Imatron), which was array elements involved increases relative to the other. Dynamic more practical, but still has limited use. The development of con- receive focusing relates to the returning ‘echoes’ reflected off temporary cardiac CT scanners allows for 0.33 s scan of the heart. objects during scanning. See Dynamic transmit focusing for Related Article: Electron beam CT methods applied to the outgoing ‘transmitted’ signal. Further Reading: Hendee, W. R. and E. R. Ritenour. 2002. Similarly to dynamic transmission focusing, the adjustment to Medical Imaging Physics, Wiley-Liss, New York. allow dynamic receive focusing is implemented electronically at the crystal array. To ensure the transducer is sensitive at the speci- Dynamic susceptibility contrast MRI fied focal area/point it must receive echoes from the focal region (Magnetic Resonance) Magnetic resonance imaging (MRI) mea- to all elements before summing them together. To achieve this surements of microcirculatory parameters of the brain, such as it implements a ‘time delay’ to account for the greater distance cerebral blood flow (CBF), cerebral blood volume (CBV) and echoes travel from the focal point to reach the peripheral elements mean transit time (MTT) add significant value to the functional D in the crystal array compared with the central ones. As the figure diagnostics of many common diseases. Perfusion MRI is of poten- below illustrates, length E2 > E1 and therefore echoes whose path tial relevance in, e.g. studies of dementia and trauma, assessment follows line E1 will receive a calculated ‘time delay’. The cumula- of the ischemic penumbra zone in acute ischemic stroke (in com- tive effect is all the echoes from the focal area are stronger and bination with diffusion MRI) to facilitate treatment decisions, in-phase compared with weak signals (acoustic noise) received investigations of intracranial vascular malformation in connec- from elsewhere. tion with neurointerventional procedures, preoperative classifi- While the delay timings relate to the focal depth specified by cation and grading of brain tumours and monitoring of various the operator, the adjustments are done automatically by the ultra- kinds of cancer therapy. sound system. The most common perfusion MRI method in clinical environ- ments is dynamic susceptibility contrast MRI (DSC-MRI). The DSC-MRI technique yields parametric maps of regional CBF, CBV and MTT and is based on the theory of intravascular or non- diffusible tracers. DSC-MRI requires rapid intravenous injec- tion of a gadolinium-based contrast agent, and after injection the tracer concentration in a tissue-feeding artery as well as in the tissue of interest is monitored by rapid imaging during the first passage of the contrast-agent bolus. The arterial concentration- versus-time curve Cart(t) is often referred to as the arterial input function (AIF). The dynamic imaging (Figure D.87) is typically accomplished by gradient-echo (GE) or spin-echo (SE) echo-planar imaging (EPI) in multi-slice mode. The employed pulse sequence must enable a temporal resolution of approximately 1.5 s, and the echo time (TE) must be sufficiently long for the contrast agent to cause a significant signal drop at peak concentration. For GE-EPI, Dynamic spatial reconstructor the temporary signal loss during the bolus passage is due to the (Diagnostic Radiology) During the early ages (1970s) of CT a contrast-agent-induced shortening of tissue T2* (i.e. a local sus- major limitation for scanning dynamic objects was the slow speed ceptibility effect). The signal-reduction effects for SE-EPI appear of gathering information for one image. to be somewhat more complex, including T2 shortening of capil- The gantry (with x-ray tube and detectors) of the second-gen- lary blood in combination with tissue-water diffusion in the local eration CT scanners was doing a 360° rotation around the patient magnetic-field gradients caused by the contrast agent. for about 1 min, later the third-generation CT scanners minimised The relationship between the image signal S and the measured this time to several seconds, but still organs such as the heart tracer concentration C is based on the assumption that C is pro- could not be scanned. This was the reason for new types of CT portional to the change in the transversal relaxation rate, i.e. C = scanners to be designed – ultra fast CT. kΔR2*. Consequently, C is calculated as C(t) = (−k/TE)ln[S(t)/S0], The first such scanner, developed by a group in the Mayo where t is time and S0 is the baseline signal. In order to obtain Clinic in the United States, was the dynamic spatial reconstructor. whole-blood volume in units of ml/100 g, a correction factor kH is This scanner used a gantry with an arc of 28 x-ray tubes. Against applied to account for brain density and haematocrit differences each tube is a detector (fluorescent type with photomultiplier). between large and small vessels, i.e. kH = 100[1 − Hlarge]/[ρ(1 − The tubes cover about 180 angular degrees around the patient. Hsmall)], where Hlarge and Hsmall are the haematocrit values in large This way a 180° scan can be made without any movement, but and small vessels, respectively, and ρ is the brain density. The only by consecutively switching the x-ray tubes. CBV is given by the time integral of kHC(t), normalised to the Additionally the whole gantry rotates slowly around the patient time integral of the measured arterial concentration Cart(t), i.e. (about 15 revolutions per minute) to cover angles bigger than 180°. During this scanning the x-ray tubes fire short x-ray pulses (sev- ¥ ò C(t)dt eral ms each). These consecutive x-ray pulses are synchronised CBV = k 0 H with the rotation, thus minimising the effective time for rotation ¥ ò Cart (t)dt and creating conditions for collecting the necessary scanning data 0 Dynamic transmit focusing 302 Dynamic transmit focusing One voxel (grey matter): 800 600 400 200 0 0 20 40 60 80 Position (x, y) = 77.33 (a) (b) D FIGURE D.87 (a) DSC-MRI time series of images obtained with a temporal resolution of approximately 1.6 s after intravenous injection of Gd-based contrast agent. (b) Typical signal curve from one single voxel in grey matter. The measured tissue concentration of tracer (corrected by the fac- by the fact that contrast-agent relaxivity appears to vary with ves- tor kH) is given by the convolution of CBF · R(t) with the AIF. R(t) sel size and geometry. An additional inherent problem is that arte- is the impulse tissue residue function, i.e. the fraction of tracer rial dispersion prohibits one single AIF to accurately represent still present in the capillary system at time t after an assumed the entire brain, i.e. the deconvolution should ideally employ the instantaneous input of tracer, i.e. AIF of the artery that actually feeds each local capillary segment. Related Articles: Perfusion imaging, Arterial input function, k Mean transit time, Cerebral blood flow, Cerebral blood volume HC(t) = CBF éëR(t) ÄCart (t)ùû Further Readings: Østergaard, L., R. M. Weisskoff, D. A. CBF can thus be obtained by deconvolution of the tissue concen- Chesler, C. Gyldensted and B. R. Rosen. 1996. High resolu- tration curve with the AIF. Finally, the mean transit time is given tion measurement of cerebral blood flow using intravascular by MTT = CBV/CBF, according to the central volume theorem. tracer bolus passages. Part I: Mathematical approach and sta- According to Zierler’s area-to-height relation, MTT can also be tistical analysis. Magn. Reson. Med. 36:715–725; Rempp, K. calculated as MTT = ∫R(t)dt/Rmax, where Rmax is the peak value A., G. Brix, F. Wenz, C. R. Becker, F. Gückel and W. J. Lorenz. of the R(t) function retrieved by deconvolution. Examples of 1994. Quantification of regional cerebral blood flow and volume parametric perfusion maps from a normal volunteer are given in with dynamic susceptibility contrast-enhanced MR imaging. Figure D.88. Radiology 193:637–641. Although DSC-MRI has become a useful tool for assessment of focal perfusion deficits, quantification is hampered by, e.g. Dynamic transmit focusing blood-brain barrier leakage, limited deconvolution performance (Ultrasound) An ultrasound beam can be focused during either in the presence of noise, arterial partial-volume effects, arterial signal transmission or receiving, or both. As the system transi- signal saturation at peak concentration, arterial signal displace- tions between transmission and receiving the proportion of ment due to local geometric distortion at peak concentration and crystal array elements involved increases relative to the other. Dynamic transmit focusing relates to the outgoing ‘transmitted’ pulse signal. See Dynamic receive focusing for methods applied to the echo ‘returned’ signal. rCBV rCBF MTT Optimum imaging, particularly of closely spaced targets, occurs with a narrow ultrasound beam. The depth of this nar- rowest point is determined in transmission by the operator focal setting – usually indicated by an arrow, triangle or circle along the side of the ultrasound image. The focus setting is most effec- tive when using a short focal length (in relation to the near field length) because the beam converges rapidly. Consequentially and detrimentally, however, the beam also diverges rapidly beyond the focal point. To achieve a narrow beam at the desired depth, the ultra- sound system must time-adjust the pulse transmission sequence from across the crystal array. This is to account for the greater distance pulses generated from the peripheral elements in the crystal array will travel compared to those from central ele- ments. As the figure below illustrates, length P2 > P1 and there- fore pulses closer in position to P2 will receive a calculated FIGURE D.88 Maps of cerebral perfusion parameters in two slices in ‘head-start’ during transmission to ensure all pulses arrive at a normal volunteer. From left to right: Colour-coded regional cerebral the focal point at the same time. The cumulative effect is all the blood volume (rCBV), colour-coded regional cerebral blood flow (rCBF) individual pulses arrive simultaneously and in-phase, achieving and mean transit time (MTT). a strong, narrow signal. Dynamic wedge 303 D ynode The central axis PDD may be measured by integrating the dose at each point during the entire irradiation of the dynamic wedge field. The central axis wedge transmission factors are determined by taking the ratio of the collected ionisation at a specified depth for the dynamic wedge field to the collected ionisation at the same specified depth for the open field with the same collimator and MU settings. It is important to note that the central axis wedge transmission factors for dynamic wedges may have much larger field size dependence than physical wedges and the field size dependence for dynamic wedges may not be asymptotic. Some manufacturers’ implementations of the dynamic wedge technique show a significant change in the trend of the central axis wedge transmission factor as the field width changes between 9.5 and 10 Dynamic wedge cm. This change in the central axis wedge transmission ratio has D (Radiotherapy) A dynamic wedge is a technique used to create been demonstrated to

approach 20%. This characteristic should an effective wedged field using independent collimator jaws. This be carefully investigated on each machine. is achieved by moving one jaw across the field during beam on, Dynamic wedge transverse beam profiles can be measured whilst keeping the other stationary, so that different parts of the with a detector array or an integrating dosimeter such as radio- field are irradiated for different lengths of time. The advantage of chromic film. When a detector array is used, the sensitivity of this method is that the beam does not undergo any beam harden- each detector must be determined. ing that would otherwise be associated with a physically wedged Dynamic wedges are offered for Varian linear accelerators field. There is also no need for physical placement of heavy (enhanced dynamic wedge, [EDW]), and Siemens linear accel- wedges within the path of the beam. erators (virtual wedge, [VW]). A wedged field of any desired angle can be created by com- Abbreviations: EDW = Enhanced dynamic wedge, PDD = bining this dynamic field with an open field of varying duration. Percentage depth dose, STT = Segment treatment table and VW Segmented treatment tables, STTs, are used to define the exact = Virtual wedge. collimator position as a function of cumulated beam monitor Related Articles: Wedge field, Dynamic jaw collimation, units, as needed to create each wedged field. A typical STT is Asymmetric jaws illustrated graphically in Figure D.89. It is usual for the upper Further Readings: Metcalfe, P., T. Kron and P. Hoban. 1997. collimator jaws to be used for this purpose as they are closer The Physics of Radiotherapy X-Rays from Linear Accelerators, to the source and hence do not need to be physically moved as Medical Physics Publishing, Madison, WI; Podgorsak, E. B. great a distance than the lower jaws. The length of treatment is 2003. Review of Radiation Oncology Physics: A Handbook for restricted by the maximum speed of movement of the collima- Teachers and Students, International Atomic Energy Agency, tor jaws. Vienna, Austria. Clinical implementation of dynamic wedges requires mea- surement of central axis PDDs, central axis wedge transmission Dynode factors and transverse beam profiles of the dynamic wedges. (General) Dynodes are a part of the photomultiplier tube (PM These measurements are complicated by the modulation of the tube) signal enhancer system. A set of dynodes (9–12) are located photon fluence during the delivery of the radiation field. in a vacuum tube following the photocathode. Photoelectrons are released from interactions in the photo-emissive substance on the entrance window. The photoelectrons are accelerated by the electric field and focused on the first dynode using a focusing grid. The dynode is coated with a substance with similar prop- erties to the photo-emissive substance used in the photocathode. 100 When a high-speed photoelectron ‘hits’ the dynode several sec- ondary electrons are released. The number of electrons released depends on the energy of the photoelectron, i.e. proportional to the voltage difference between the photocathode and the dynode. The electron multiplication process is repeated when the second- ary electrons are accelerated against the second dynode which is maintained at a higher potential than the first dynode. The 20 process is repeated for all dynodes leading to a ‘shower’ of elec- trons reaching the anode. The total signal multiplication factor depends on individual dynode multiplication factors and the num- –10 10 ber of dynodes. Relatively low stimuli can therefore be enhanced Jaw position (cm) into a large detectable pulse using a PM tube. Related Articles: Photomultiplier (PM) tube, Photocathode of FIGURE D.89 A typical STT to create a wedged field. An open field photomultiplier tube is delivered for the first 20% MU, after which one jaw starts moving Further Reading: Cherry, S. R., J. A. Sorenson and M. E. continuously across the field, stopping 5 mm away from the stationary Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, diaphragm. Philadelphia, PA, pp. 101–102. Cumulate monitor units (% ) E Ear protection introduction of new fast spin-echo sequences during which the (Magnetic Resonance) During the image acquisition a significant gradient amplitudes increase and the rise times decrease. amount of acoustic noise is produced by vibrations of the gradient Acoustic noise levels for fast MRI pulse sequences have been coils which are embedded in solid material around the bore of the measured on MRI systems with field strengths ranging from 0.2 magnet. The electrical current passing through the gradient coils to 3 T. Measured noise levels varied from 82.5 dB(A) for a 0.23 T produces a large force which tends to move the coils and causes system to 118.4 dB(A) for a 3 T system. vibration, thus generating loud sounds such as tones, knocking or dB(A) or A-weighted decibels are decibels with the sound banging noises associated with the frequency and waveform of pressure scale adjusted to conform to the frequency response of the applied gradient pulse. the human ear. The intensity level or relative level of a noise is its intensity Analysis of the measurements showed that relative to an agreed ‘zero’ intensity. The latter is chosen as the noise intensity I0 at the threshold of hearing which is generally E 1. Pulse sequence parameters, particularly FOV and TR, accepted as 10−12 W/m2 (1 pW/m2) at 1 kHz. Since the ear has a were more influential in determining noise level than logarithmic response to noise intensity, it is convenient to define field strength the unit of intensity level using a logarithmic scale. 2. The noise level was found to vary along the z-direction The intensity level of a sound of intensity I (W/m2) measured with a maximum near the bore entrance in bels (B) is then defined by 3. In some cases there was a significant increase in noise with a volunteer present instead of a test object æ I ö Intensity level = log10 ç ÷B è I Acoustic noises up to 65 dB are acceptable while between 65 and 0 ø 95 dB the noise is unpleasant but not directly harmful. Above 95 dB hearing damage can occur. The bell is a large unit representing intensity differences in the Noise is a complex disturbance having many component fre- ratio 10:1 and usually it is more common to use a smaller unit, the quencies and producing varying effects on exposed individuals. decibel (dB), given by High intensity noise is generally accepted as greater than 85 dB and as the noise level increases, there are a number of harmful æ I ö Intensity level = 10 log results: 10 ç ÷dB è I0 ø • Anxiety, headaches and temporary hearing losses where 1 B = 10 dB. • Change in hearing acuity and possible damage to the In the following table intensity levels are reported for different cochlea sources of noise with the indication of the associated average and • Stimulation of receptors in the skin peak pressure. • Significant changes in pulse rate • Vibration of muscles and reduced coordination • Feeling of fear, annoyance, dissatisfaction Sound Average Peak • Nausea, vomiting, dizziness (>30 dB) Intensity Intensity Pressure Pressure • Pain in the middle ear (≈140 dB) Source of Noise Level (dB) (W/m2) (Pa) (Pa) Temporary hearing loss has been also reported using conven- Threshold of 0 (silence) 10−12 0.00002 0.00003 tional pulse sequences. Gradient noise may also interfere with hearing at 1 kHz patient communication. Library 20 10−10 0.0002 0.0003 The International Electrotechnical Commission (IEC) advises Background 40 10−8 0.002 0.003 MR manufacturers to provide a warning if the acoustic noise from music a given pulse sequence is likely to exceed 99 dB(A), which is suf- Speech at 0.6 m 60 10−6 0.02 0.03 ficiently removed from 140 dB(A), the level at which damage to Heavy traffic 80 10−4 0.2 0.3 the hearing may occur. Jet overhead 100 10−2 2 3 An acceptable and inexpensive means for the prevention of Threshold of 120 1 20 30 hearing loss is the regular use of disposable earplugs for patients feeling undergoing MR examinations. They can decrease the noise intensity by about 30 dB. A more expensive alternative would be ‘antinoise’ or destructive noise apparatus which, not only reduces The decibel level of noise within the gantry of the MR scan- noise, but also permits a better communication between the ner varies from about 82 dB to about 120 dB, and its intensity operator and the patient during the examination. They consist of level depends on the pulse sequence used for the imaging proce- devices performing a real-time Fourier analysis of the noise pro- dure. Recently noise levels have increased considerably with the duced by the gradient coils and generating a signal with the same 305 Earth’s magnetic field 306 Echo planar imaging (EPI) physical characteristics but opposite phase which is added to the with short phase encoding gradient pulses (Gy). The sequence dia- original noise. The subsequent cancellation of the repetitive noise gram and corresponding k-space for a standard gradient-echo EPI permits a reduction of its intensity level of 50%–70%. sequence are shown in Figures E.1 and E.2, respectively. Each MR operators and accompanying family members will suffer echo is generated by the refocusing action of the readout gradi- the same noise exposure if they remain in the examination room ents, and the effective echo time (TE) is defined as the time from during the image acquisition and therefore they should use the the RF pulse to the central echo (ky = 0). same hearing protection as the patients. EPI can also be employed as a spin-echo experiment by adding The following table shows the relationship between the noise a 180°-pulse after the 90°-pulse shown in the following. Another duration and recommended permissible sound levels for occupa- variation of the EPI sequence is spiral imaging with two oscillat- tional exposures. ing gradients, generating a spiral k-space trajectory. EPI is very demanding on the imaging hardware because large field gradients have to be generated and switched rapidly. Another drawback is its sensitivity to magnetic field inhomogeneities pres- Noise Duration per Day (h) Sound Level [dB(A)] ent in boundaries between tissue and air. This behaviour arises 8.0 90 from the low bandwidth in phase encoding direction and the 6.0 92 rather long effective TE, leading to distortion and in severe cases 4.0 95 to gross signal dropout. 3.0 97 E 1.5 100 1.0 102 90° 0.5 105 0.25 115 rf These noise levels are based upon a long-term exposure to noise. The levels for chronic exposure concern only those MR operators Gz who are constantly working in a noisy environment. Further Readings: Miller, L. E. and A. M. Keller. 1972. Regulation of occupational noise. In: Handbook of Noise Control, ed., C. M. Harris, McGraw-Hill, New York, pp. 1–16; Price, D. Gy L., J. P. De Wilde, A. M. Papadaki, J. S. Curran and R. I. Kitney. 2001. Investigation of acoustic noise on 15 MRI scanners from 0.2 T to 3 T. J. Magn. Reson. Imaging 13:288–293. Gx Earth’s magnetic field (Magnetic Resonance) The Earth’s magnetic field is approxi- mately the field of a dipole positioned at the centre of the Earth. The dipole has a declination by approximately 11.3° from the ADC planet’s axis of rotation. The flux density of the magnetic field at the Earth’s surface ranges from some less than 30 μT (0.3 G) to FIGURE E.1 EPI pulse sequence. over 60 μT (0.6 G) around the magnetic poles. The Earth’s field is changing in size and position. The magnetic poles may shift as much as 15 km a year. ky Earthing (General) Earthing is a term used in UK electrical engineering to represent electrical equipment that has direct physical connection to Earth thus providing protection against electrical shock. In the United States the equivalent term is ‘grounding’. Echo enhancing agent (Ultrasound) Echo enhancing agents are used as contrast media in ultrasound examinations. These consist of microbubbles, i.e. kx micron-sized bubbles, stabilised by a shell. Further details are found under Contrast agents. Echo planar imaging (EPI) (Magnetic Resonance) Echo planar imaging (EPI), proposed by Mansfield in 1977, refers to the concept of collecting all phase encoding steps needed to reconstruct an image using one series of echoes after a single RF excitation. Following the RF excita- tion, the readout gradient (Gx) is rapidly reversed in combination FIGURE E.2 Traversal of

k-space in EPI. Echo-planar imaging and signal targeting 307 Echo time (TE) The speed at which images can be obtained with EPI has Wong, B. Siewert, S. Warach and R. R. Edelman. 1998. A gen- allowed several applications for dynamic studies of the brain eral kinetic model for quantitative perfusion imaging with arterial such as functional MRI (fMRI), diffusion imaging and perfusion spin labeling. Magn. Reson. Med. 40:383–396; Edelman, R. R., B. imaging. Siewert, D. G. Darby, V. Thangaraj, A. C. Nobre, M. M. Mesulam Related Articles: Diffusion imaging, Echo spacing, fMRI, and S. Warach. 1994. Qualitative mapping of cerebral blood Gradient echo, Perfusion imaging, Spin echo flow and functional localization with echo-planar MR imaging Further Reading: Haacke, E. M., R. W. Brown, M. R. and signal targeting with alternating radio frequency. Radiology Thompson and R. Venkatesan. 1999. Magnetic Resonance 192:513–520; Hacke, E. M., R. W. Brown, M. R. Thompson and Imaging, Physical Principles and Sequence Design, John Wiley R. Venkatesan. 1999. Magnetic Resonance Imaging: Physical & Sons, Inc., New York. Principles and Sequence Design, John Wiley & Sons, Inc., New York. Echo-planar imaging and signal targeting with alternating radiofrequency (EPISTAR) Echo ranging (Magnetic Resonance) The EPISTAR arterial spin labelling (Ultrasound) Echo ranging is a distance measurement performed (ASL) preparation method is based on the subtraction of two by measuring the time interval between the transmission of a images, one acquired with a preparatory inversion of inflowing sound pulse and the return of its echo. See also Pulse echo. arterial spins (the labelling experiment) and one without inversion (the control experiment). For the inversion of arterial spins, a spin Echo spacing tagging radio-frequency (RF) pulse (normally a 180° RF pulse) is (Magnetic Resonance) Echo spacing refers to the distance in E applied in an axial slab proximal (caudal) to the imaging plane. In time between the echoes in pulse sequences employing multiple the control experiment, an inversion pulse is applied to a slab dis- echoes, e.g. echo planar imaging and fast spin echo. The echo tal (superior) to the imaging slice in order to cancel magnetisation spacing affects the amount of image artefacts and the maximum transfer effects. After a time delay TI the image is acquired with a number of slices per repetition time. fast read-out sequence such as EPI, spiral reconstruction, HASTE Related Articles: Echo planar imaging, Fast spin echo or 3D GRASE. One or more presaturation pulses are applied immediately before the inversion pulse to suppress signal from the Echo time (TE) static tissue. The inverted arterial spins, in labelling experiment, (Magnetic Resonance) Echo time, TE, is the time from the RF cause a degeneration of the MR signal in the imaging slab. The excitation to the centre of the echo being received. Together with degree of the MR signal depression corresponds to the amount of repetition time (TR) and excitation pulse flip angle, TE is an the regional perfusion and the subtracted image can be viewed as important parameter in the build-up of image contrast. TE deter- a qualitative perfusion-weighted image. By fitting the subtracted mines how much decay of the transverse magnetisation is allowed image to a model, usually used is the general kinetic model devel- to occur before the signal is read. Short TE allows less T2 or T2* oped by Buxton et al. (1998) and quantitative perfusion maps can signal decay while a longer TE leads to a smaller signal. A long be obtained. The major disadvantages of this method are: the TE in combination with a long TR produces T2-weighted images. compensation for magnetisation transfer effects is complete only A short TE together with short TR leads to T1-weighted images. A half way between the inversion slabs which makes method suited proton-weighted image is obtained when a short TE in combina- only for single slice imaging and the blood, inverted by a control tion with a long TR is used (Figure E.4). pulse, enters into the imaging slice from the distal side and affects TE is often given in milliseconds. the subtracted image negatively (Figure E.3). Related Articles: Effective echo time, Flip angle, Repetition Related Articles: Perfusion imaging, Arterial spin labelling, time, TR, T1-weighted, T2-weighted FAIR, PICORE, QUIPSS – QUIPSS II – Q2TIPS Further Readings: Bernstein, M. A., K. F. King and x. J. Zhou. 2004. Handbook of MRI Pulse Sequences, Elsevier Inc., TR Amsterdam, the Netherlands; Buxton, R. B., L. R. Frank, E. C. TE 90° 180° 90° RF Inversion slab Imaging GS slice GP GF (a) (b) Uninverted inflowing spins Uninverted static spins ADC Inverted inflowing spins FIGURE E.4 Spin-echo pulse sequence. RF and ADC are the radio fre- FIGURE E.3 The principle of an EPISTAR pulsed ASL labelling tech- quency pulse and the signal received from the slice. GS, GP and GF are nique. (a) Illustrates the labelling experiment and (b) shows the corre- the slice selective, phase and frequency encoding gradients. TE is the time sponding control experiment. between the 90 pulse and the ADC. Echo train length (ETL) 308 Eddy currents Echo train length (ETL) (Magnetic Resonance) Echo train length is the number of echoes created after one excitation. In particular, this refers to the num- ber of applied 180° rephasing pulses and their corresponding echoes during one repetition time (TR) for a fast spin echo pulse sequence type. Another name for echo train length is turbo factor. See Fast spin echo (FSE) for more information. Related Articles: Fast spin echo (FSE), Half acquisition sin- gle-shot turbo spin echo (HASTE), Rapid acquisition relaxation enhancement (RARE), Turbo spin echo (TSE) Echocardiography (Ultrasound) Echocardiography refers to ultrasound examina- tion of the heart. The echocardiogram gives information as to the structure and motion of the heart and the blood flow within it. A typical cardiac examination may use many of ultrasound’s modalities; M-mode (Figure E.5), B-mode, colour flow imaging and pulsed and continuous wave spectral Doppler. Recent devel- FIGURE E.6 The position on the chest wall which will produce a long E opments in echocardiography include 3 and 4D imaging, contrast axis plane image of the heart. agent imaging and tissue Doppler imaging. The examination includes the pericardium (the tissue surrounding the heart), the structure of the heart including the chambers and valves and the velocity of blood flow in the heart. Echocardiography is used to examine many cardiac conditions including heart valve disease, ischaemic heart disease, cardiomy- opathies, pericardial disease and cardiac masses. It is used to examine the great vessels, in pulmonary disease, in arrhythmias and in the assessment of left ventricular function. The approach may be transthoracic which is performed through the chest wall (Figure E.6) or transesophageal where the trans- ducer is passed endoscopically into the esophagus. The examina- tion can be combined with exercise – a stress echocardiogram. Related Article: Stress echocardiography Echogenic (Ultrasound) Echogenic, or hyperechoic, describes echoes which are unusually bright. An example is shown in Figure E.7 where the cortex and medulla appear markedly brighter than is usually seen. The kidney can be described as echogenic. FIGURE E.7 Longitudinal image of a kidney. The appearance shows Eddy currents bright echoes in the cortex and medulla and, compared with adjacent (Magnetic Resonance) Electric currents can be induced in a structures and the normal appearance of kidneys, can be described as conductor due to changes in magnetic field or by motion of the echogenic. conductor through a magnetic field. These electric currents are called eddy currents. When the magnetic field is changed it leads to changes in flux, Φ, and according to Faraday’s law of induction, the change in flux through an area, Ω, creates an electromotive force (emf) in the conducting material: ¶F ¶B emf = - = dA ¶t ò ¶t W where B is the magnetic field dA is a small part of the area Ω The main source of eddy currents is the rapidly varying gra- dient fields. When the gradients are switched on and off, the change in magnetic field induces a current in the conductor in the FIGURE E.5 An M-mode line and M-mode display across the aortic opposite direction to counteract the change in magnetic field (see root and left atrium. Figure E.8). Edge artefact 309 E dge illumination G(t) Geddy(t) t t FIGURE E.8 When a gradient (a) is switched on, the positive slope will induce a negative eddy current (b), and when the gradient is switched off, the negative slope will induce a positive eddy current. In the MR environment the eddy currents are induced by largest differences in pixel values from their local neighbours. For the gradient fields in all conducting material in the surround- example, consider the following mask: E ing environment (e.g. the cryostat). The currents can introduce artefacts to the images and seriously degrade overall magnetic -1 -1 -1 performance. Shielding of gradient coils as well as inclusion of eddy current compensation in the gradient pulse design -1 8 -1 are frequently used methods to reduce eddy current effects -1 -1 -1 (Figure E.9). In the right image strong diffusion gradients (b = 800 s/mm2) When convolving an image with this mask, the results will be result in distortions. zero for all pixels whose neighbours have the same value but dif- fer from zero if there is a local difference in pixel values. An edge Edge artefact in an image will then be intensified when applying this filter. The (Ultrasound) See Edge shadow filter can be modified to enhance edges in a particular direction. Examples are the Prewitt operator: Edge detection (Nuclear Medicine) Edge detection is the process of identifying -1 -1 -1 -1 0 -1 organ or organ substructure boundaries in morphological images. 0 0 0 and -1 0 -1 Edge enhancement -1 -1 -1 -1 0 -1 (General) In image processing it is sometimes of importance to enhance local changes in count density to visualise edges or Edge illumination boundaries. For example, the volume of an organ may need to (Diagnostic Radiology) Edge-illumination (EI) imaging is a be calculated and in order to do this the boundary of the organ phase-contrast imaging technique which was developed as an needs to be determined. It is necessary to use an imaging modal- alternative method to the analyser-based imaging (ABI), where ity that shows a contrast difference between the organ of interest the analyser crystal is replaced by one or (more commonly) two and the background. The method identifies those pixels that have absorbing slits or masks. Not implying the use of a perfect crystal, EI allows a relaxation of the requirements of mechanical stability and beam coherence, making the technique applicable to table- top set-ups employing x-ray tubes. As in the case of ABI, the signal formation in EI is based on the detection of the x-ray refraction angle due to the presence of a sample. The working principle of EI in its simplest synchrotron-based implementation is sketched in Figure E.10 The x-ray beam is collimated down to a narrow blade by an absorbing slit placed upstream from the sample (the pre-sample slit) featuring an aper- ture of a few to tens of micrometres. A second slit (the detec- tor slit) is placed in front of the detector and aligned with a row of pixels. The two slits are slightly misaligned so that the beam collimated by the pre-sample slit gets partially absorbed by the detector slit and partially collected by the detector pixels. When a refractive object is put downstream from the pre-sample slit, FIGURE E.9 the x-ray beam is refracted hence its (average) direction is shifted Eddy currents can distort diffusion weighted MR images, which utilise strong magnetic gradients. The white outline illustrates the either towards the detector pixels or towards the absorbing slit, position of the edge of the image on the left without diffusion gradients resulting in an increase or decrease of the detected intensity. (b = 0 s/mm2). Considering the described set-up, EI is sensitive to the component Edge illumination 310 Edge illumination FIGURE E.10 Sketch of the original EI implementation. The object is scanned through the aperture of the pre-sample slit. (a) Photons refracted onto the detector slit producing a signal decrease, (b) photons traversing the sample far from sharp interfaces are mostly non-refracted and (c) photons are refracted onto the detector producing a signal increase. (d) Shows the final image where the

vertical dimension encodes the scanning position. E of the refraction angle which is perpendicular with respect to the can be acquired and processed to yield, separately, absorption, aperture direction. refraction (i.e. differential phase) and ultra-small-angle scatter- Being the refraction angle proportional to the gradient of the ing (USAXS, sometimes referred simply to as scattering or dark- sample-induced phase shift, EI falls within the category of the field) maps. differential phase-contrast techniques. When a broad beam is used to irradiate the object (e.g. when A bi-dimensional image can be obtained by scanning the using an x-ray tube), the EI configuration can be replicated over sample through the beam and recording the detected intensity as a wider field of view by replacing the slits with absorbing masks a function of the scanning position. During the scan, the boundar- featuring many apertures, each producing a narrow ‘beamlet’. ies of the various object structures, where refraction mainly takes The pitches of pre-sample and detector masks are usually chosen place, will produce positive or negative signal peaks which are to match (accounting for the beam divergence) the pixel size, so analogous to the ABI image recorded at one slope of the rocking that every row or column of pixels corresponds to an aperture (see curve. Figure E.11b). In principle this arrangement makes the sample The analogy between ABI and EI can be further exploited by scan unnecessary, thus speeding up the acquisitions as different noting that varying the misalignment between pre-sample and portions of the sample are imaged at the same time. detector slits is conceptually equivalent to changing the angular As a side remark, edge illumination is sometimes confused displacement of the analyser crystal. In fact, when the slits are with the grating interferometry (GI) technique, and vice versa, perfectly aligned the whole (collimated) beam is collected by the given the apparent similarity between the experimental set-ups. pixel whereas, as the misalignment increases, larger fractions of Actually the two techniques are intrinsically different: in EI each the beam get absorbed by the detector slit, hence the detected beamlet produced by the pre-sample mask is analysed indepen- intensity is reduced. The curve expressing the detected intensity dently from the others and no interference between different as a function of the misalignment between the slits is called the beamlets occurs, whereas GI is based on the interference pattern illumination curve (see Figure E.11a). As in the case of ABI, sev- of the periodically modulated x-ray radiation field produced by eral images at different working points on the illumination curve the phase grating. FIGURE E.11 (a) Illumination curve, that is the detected intensity by one detector pixel as a function of the misalignment between the pre-sample and detector slits or masks. (b) Diagram of a typical laboratory EI set-up based on the use of an x-ray tube and two masks (not to scale). Edge shadow 311 Editing, spectral Related Articles: Phase-contrast imaging, Analyser-based Edge spread function imaging, Grating interferometry (Diagnostic Radiology) The edge spread function (ESF) is a sys- Further Readings: Diemoz, P.C. et al. 2017. Non- tem’s response to the acquisition of an ‘ideal’ edge (one that is interferometric techniques for x-ray phase-contrast biomedi- perfectly straight and of infinite contrast). The ESF is obtained by cal imaging. In: Handbook of X-ray Imaging: Physics and imaging an edge test object (usually a precisely machined sheet Technology. ed. Russo, P., CRC Press, pp. 999–1023; Endrizzi, of metal) and placing a region of interest (ROI) along the imaged M. 2018. X-ray phase-contrast imaging. Nucl. Instrum. Methods edge (once the image has been linearised). Each pixel value in Phys. Res. Section A: Accelerators, Spectrometers, Detectors and the ROI is then projected onto a line perpendicular to the edge Associated Equipment 878:88–98; Olivo, A. and E. Castelli. 2014. (Figure E.12). X-ray phase contrast imaging: From synchrotrons to conventional As the edge spread function can be derived as the integration sources. Riv. del nuovo cimento 37(9):467–508. of a large number of differential elements which behave as line sources, the first derivative of the ESF will give the line spread Edge shadow function (LSF) of the system. The modulation transfer func- (Ultrasound) Edge shadow is an artefact typically characterised tion (MTF) of the system can then be obtained through a dis- as a pair of dark vertical streaks beneath the lateral (side) edges of crete Fourier transform of the LSF. The edge test object must be cystic regions. This is proposed to be caused by refraction. imaged at a slight angle to the detector pixels to obtain an over- Refraction of the ultrasound beam occurs when it obliquely sampled ESF. This ensures the MTF can be measured beyond the (one that is not parallel or perpendicular to axial beam direc- Nyquist limit. tion) incidents an interface surface of a medium with a different Related Articles: Line spread function (LSF), Point spread E speed of sound to the one it is currently traversing. The lateral function (PSF), Modulation transfer function (MTF) width of the beam and oblique angle of incidence means that Further Reading: International Standard IEC 62220-1; Kohn, one side of the beam will enter the new medium before the other K. Modulation transfer function measurement method and results side. Depending on whether the new medium’s speed of sound is for the Orbview-3 high resolution imaging satellite. greater or less than the previous medium, the leading side of the beam will increase or decrease its transmission speeds relative to Editing, spectral the other side causing the beam to change its direction of travel (Magnetic Resonance) Spectral editing refers to a range of tech- slightly. The degree of change is dependent on the angle of inci- niques employed to simplify NMR spectra and so facilitate inter- dence, relative difference between the speed of sound properties pretation and quantification, particularly when the appearance of of the two mediums and the beam width. spectra is complicated by overlapping lines. The crystal elements in the transducer array will assume that Water suppression is a common example of spectral editing. all echoes are received from a non-deviated axis (i.e. a vertical More generally, the following are examples of spectral editing column directly below it). Echoes received from a refracted beam technique. will therefore be misplaced on the image. This artefact is commonly seen where the speed of sound is less within the cyst compared to the surrounding medium. Where a beam incidences on the upper right surface the left side will slow first, pulling the beam to the left. The severity of this pull to the left increases for beams interfacing further towards the cyst’s lateral edge causing a dearth of echo from below this region. This is shown in the figure. The opposite occurs on the other side. Edge spread (Nuclear Medicine) The edge-spread function (ESF) is closely related to the point-spread function (PSF) and line-spread func- tion. An edge is imaged and the spatial resolution and sharpness degradation is evaluated by defining a profile through it. The ESF is then used to calculate the modulation transfer function (MTF) which is a measure that describes the degradation of an imaging system in terms of frequencies and amplitudes. Abbreviations: Edge-spread function, LSF = Line-spread FIGURE E.12 Pixel values along the image of an edge object are pro- function and PSF = Point-spread function. jected onto a line perpendicular to the edge to obtain a system’s edge Related Article: Modulation transfer function spread function. Effective dose (E) 312 E ffective dose 1. Relaxation-based techniques: When a spectrum con- The starting point for determining effective dose is the tains one or more overlapping lines with different T1 absorbed dose. Absorbed dose modified by radiation weighting relaxation times, a broadband inversion pulse followed factor, wR, produces the equivalent dose, HT. Effective dose is a by a suitable delay can be used to reduce (or ideally summation of the equivalent doses to each of the organs irradi- null) signal from one of the overlapping lines. Similarly, ated modified by their tissue weighting factor, wT. Thus a spin-echo experiment can be used when the T2 relax- ation times of the two species to the extent that it is pos- E = sible to choose an echo time such that signal from one åwT HT T species remains while that from the other has substan- tially decayed. The unit of effective dose is the Sievert (named after Rolf Sievert). 2. J-coupling–based techniques: These methods can be Tissue weighting factors are loosely based on tissue sensitivity employed when one of the overlapping species is a to ionising radiation and have been defined by ICRU (International coupled nucleus. In such a case, the signal from this Commission on Radiological Protection). The values for tissue species is modulated in amplitude and phase during the weighting factors determined by ICRP in Publication 103, 2007 echo time of a spin-echo experiment. This phenomenon are given in Table E.1. can be exploited by performing a series of experiments To indicate how the effective dose would be calculated, con- in which the echo time is changed, and in some cases sider a procedure in which an irradiation of the thorax delivered also using decoupling, followed by signal subtraction to E absorbed doses of 15 mGy to the thyroid, 10 mGy to the lung, eliminate unwanted lines and enhance others. Various 30 mGy to the breast, 45 mGy to the heart and 10 mGy to the polarisation transfer and multiple quantum filtering skin. As the absorbed dose was due to x-radiation the radiation techniques, beyond the scope of this article, may also factor, wR, is 1 and so the equivalent dose to each organ will be be considered examples of techniques in this category. numerically equal to the absorbed dose. Then using the factors 3. Two-dimensional NMR: This is a very general class in the preceding table, the effective dose can be calculated as of technique used in chemical analysis, the details of follows: which are beyond the scope of this article. Briefly, a 2D NMR experiment consists of a series of acquisitions in E = 15* 0.05 +10* 0.12 + 30* 0.12 + 45* 0.1+10* 0.1 which spins are allowed to evolve for an incremented period of time under the influence of some process or = 0.75 +1.2 + 3.6 + 4.5 +1 interaction that modulates the signal as a function of = 11.05 the evolution time. The effect of this modulation can be displayed perpendicular to the conventional frequency E = 11.05mSv axis in a two-dimensional plot, which may allow over- lapping peaks to be separated out. Note that the dose to a particular organ, e.g. the lung, is the dose A complication of spectral editing is that very often the intensities averaged over the whole organ. Partial irradiation of organs is not of the desired signals are altered in the process of suppressing the considered in the calculation of effective dose. unwanted ones, so that quantification is hampered. Related Articles: Absorbed dose, Equivalent dose, Radiation Related Articles: Decoupling, Magnetic coupling, Water weighting factors, Tissue weighting factors suppression Effective dose Effective dose (E) (Nuclear Medicine) Effective dose is a measure of dose to the (Radiation Protection) This is the radiation dose quantity in radi- whole body from the administration of a radionuclide, account- ation protection used for purposes of risk comparison. It provides ing for the biological effectiveness of the type of radiation, and an estimate of the dose to the whole body from a single or series the relative radiosensitivies of the organs irradiated. The unit of of partial body irradiations. effective dose is the Sievert (Sv). TABLE E.1 Proposed Tissue Weighting Factors per ICRP Publication 103 Tissue wT ΣwT Bone-marrow (red), Colon, Lung, Stomach, Breast, Remainder tissues* 0.12 0.72 Gonads 0.08 0.08 Bladder, Oesophagus, Liver, Thyroid 0.04 0.16 Bone surface, Brain, Salivary glands, Skin 0.01 0.04 Total 1.00 * Remainder tissues: Adrenals, Extrathoracic (ET) region, Gall bladder, Heart, Kidneys, Lymphatic nodes, Muscle, Oral mucosa, Pancreas, Prostate (#), Small intestine, Spleen, Thymus, Uterus/cervix ($). * Adrenals, extrathoracic region, gall bladder, heart, kidneys, lymphatic nodes, muscle, oral mucosa, pancreas, prostate, small intestine, spleen, thymus, uterus/cervix Effective dose equivalent 313 Effective detective quantum efficiency Effective dose can be calculated by multiplying the Effective dose equivalent absorbed dose in Gray (Gy) for each organ by a radiation (Radiation Protection) This

is the term that was formerly used weighting factor accounting for the biological damage caused to describe the effective whole body dose equivalent to a partial by the type of radiation, and a tissue weighting factor account- body exposure. This term is no longer used. ing for the radiosensitivity of the organ irradiated. Summing The current term is Effective dose. for each organ will give a whole-body effective dose (E) in Related Article: Tissue weighting factor Sieverts (Sv): Effective detective quantum efficiency E = åwT ´ DT ´ wR = åwT ´ HT (E.1) (Diagnostic Radiology) The detective quantum efficiency (DQE) T T of a detector characterises its signal to noise performance in the context of how efficiently incident photons are transferred into an Where DT is the average absorbed dose in organ T, wR is the radia- image. However, detector DQE does not reflect the contributions tion weighting factor dependent on the radiation type, wT is the of scattered radiation, the anti-scatter grid, magnification, and tissue weighting factor for organ T, and the summation is over all focal spot blur on the quality of images acquired clinically. The organs. effective DQE (eDQE), established independently alongside the The absorbed dose DT multiplied by the radiation weighting generalised DQE (gDQE), was developed as a metric to represent factor wR is also commonly defined as the equivalent dose, HT the DQE of the entire imaging system. as shown in Equation E.1. The unit of equivalent dose is also the eDQE can be obtained by measurement of the various Sievert (Sv). parameters required for a DQE calculation but with a phan- E Effective dose can therefore also be defined as the equivalent tom present to represent a patient. The eDQE is defined by the dose HT in organ T, multiplied by the tissue weighting factor for equation: organ T and summed over all organs. 2 eDQ ( ) MTF ( f ¢) × - SF ¢ (1 )2 For gamma rays, x-rays and electrons, the radiation weight- ing factor accounting for biological damage, wR = 1; for alpha E f = NNPS( f ¢) × TF × E × q particles, wR = 20. As such, the biological damage resulting from an absorbed dose delivered by alpha particles is 20 times where: that from gamma rays or electrons. These differences are due eDQE( f′) (in units of %) is the effective DQE at a magnified to the type, energy, and method of energy deposition within frequency of f′ (i.e. the frequency at the entrance plane tissue. Alpha particles have a short range in tissue, and there- of the phantom, given by f′ = mf, where m is the mag- fore deposit their energy within a localised region. In contrast, nification and f is the spatial frequency in the plane of gamma rays and electrons deposit their energy over a much the detector). larger area. MTF( f′) (no unit) is the measured MTF (using established Tissue weighting factors (wT), describing the relative radiosen- techniques) in the presence of phantom, scattered radia- sitivity of each organ for the average human, are published by tion, focal spot blur, and the anti-scatter grid. the International Commission on Radiological Protection (ICRP). SF (no unit) is the measured scatter fraction, determined Table E.1 gives some examples of tissue weighting factors for using a beam stop technique. Here an array of ‘beam common organs. stops’ are imaged in front of the phantom with scatter As shown in Equation E.1, calculating the effective dose from fraction then determined as the ratio of the exposure the administration of a radionuclide requires knowledge of the behind the beam stops to that surrounding the beam absorbed dose to each irradiated organ. In practice, this can be stops. determined from time activity curves describing the kinetics of NNPS( f′) (in units of mm2) is the measured normalised noise the radionuclide within each organ as described in the MIRD power spectrum (using established techniques) in the methodology. The absorbed dose from each organ is then simply presence of phantom, scattered radiation, focal spot multiplied by the relevant wR and wT values and summed across all blur and anti-scatter grid. organs to obtain a value of effective dose. TF (no unit) is the measured transmission fraction through the Because the effective dose is correlated with the risk of bio- phantom using a narrow beam geometry, determined as logical damage to tissue, it is a useful measure for comparing the the ratio of the exposure with and without the phantom risks associated with examinations within nuclear medicine and present. across other modalities. Despite its utility as an overall indica- E (in units of dose, e.g. mGy) is the free-in-air exposure, mea- tor of radiation risk, effective dose should however be used with sured in front of the phantom and inverse-squared cor- caution. The use of effective dose is not, for example, recom- rected to the detector plane. mended in radionuclide therapy, nor should it be used to evaluate q (in units of dose/mm2) is the ideal squared signal-to-noise- the risk from a radionuclide study to a particular individual, as ratio per unit exposure at that detector plane, deter- the calculation of effective dose is based on the average human, mined by simulation of an ideal counting detector with and actual doses can vary considerably with organ shape, size an incident primary beam attenuated by the material of and pathology. the phantom. Related Articles: Absorbed dose, equivalent dose, MIRD, tis- sue weighting factor, Radiation weighting factor Related Articles: Modulation transfer function (MTF), Noise Further Readings: Cherry, S. R., J. A. Sorenson and M. power spectrum (NPS), Detector quantum efficiency (DQE), E. Phelps. Physics in Nuclear Medicine, 4th edn.; ICRP. 2007. Generalised detective quantum efficiency (gDQE) The 2007 recommendations of the International Commission Further Reading: Samei, E. et al. 2009. Effective DQE on Radiological Protection. ICRP Publication 103. Ann. ICRP (eDQE) and speed of digital radiographic systems: An experi- 37(2–4). mental methodology. Med. Phys. 36(8):3806–3817. Effective dynamic range 314 Effective point of measurement Effective dynamic range associated with hospital light sources with reference to the Control (Magnetic Resonance) See Dynamic range of Artificial Optical Radiation at Work Regulations 2010. J. Radiol. Prot. 30(3):469; ICNIRP. A closer look at the thresholds Effective echo time of thermal damage: Workshop report by an ICNIRP task group. (Magnetic Resonance) The effective echo time, TEeff, is the time Health Phys. 111(3):300–306; ICNIRP. 2013. Guidelines on limits period between the excitation pulse and the time when the central of exposure to incoherent visible and infrared radiation. Health k-space line, corresponding to the echo without phase encoding, Phys. 105(1):74–91; ICNIRP. 2013. Guidelines on limits of expo- is acquired. See Fast spin echo (FSE) or Echo planar imaging sure to laser radiation of wavelengths between 180 nm and 1,000 (EPI) for more information. μm. Health Phys. 105(3):271–295; ICNIRP. 2004. Guidelines on Related Articles: Echo train length, Fast spin echo (FSE), limits of exposure to ultraviolet radiation of wavelengths between Echo planar imaging, Half acquisition single-shot turbo spin echo 180 nm and 400 nm (Incoherent Optical Radiation). Health Phys. (HASTE), Turbo spin echo (TSE) 87(2):171–186; ICNIRP. 2000. Revision of the guidelines on lim- its of exposure to laser radiation of wavelengths between 400 nm and 1.4 μm. Health Phys. 79(4):431–440. Effective energy (General) Effective energy is a characteristic of a polyenergetic Effective focal spot x-ray beam that compares it to a monoenergetic beam based on (Diagnostic Radiology) See Focal spot, effective penetrating capability. The effective energy of an x-ray beam can E be determined by measuring the HVL, and from that calculat- ing the effective linear attenuation coefficient μeff. The effective Effective half-life energy of the beam is the photon energy that has the same attenu- (Nuclear Medicine) Effective half-life is the time period in which ation coefficient value as determined from published references. the amount of deposited radionuclides in an organism is reduced In practice the effective energy of a moderately filtered x-ray to 50% of the initial value. The decrease in radionuclide is due to beam is approximately 2/3 of the maximum photon energy (peak two different processes – physical decay and biological excretion. kV). For example, an x-ray beam produced with 90 kVp will have The effective half-life TEff depends on the physical half-life of the an effective energy of approximately 60 keV. radioisotope TPhys and biological excretion TBio: Further Reading: Bushberg, J. T., J. A. Seibert, E. M. Leidholdt and J. M. Boone. 2002. The Essential Physics of TPhys ×T T Bio Eff = Medical Imaging, 2nd edn., Lippincott Williams & Wilkins, T Phys + TBio Philadelphia, PA. The effective half-life can also be derived from the physical and biological decay constants λPhys and λBio, respectively. The effec- Effective exposure, AORD tive decay constant λEff is the sum of the physical and biological (Non-Ionising Radiation) The term is used to indicate the light decay constants: dose received by a person which is likely to induce a biologi- cal effect. This is calculated by considering raw exposure and lEff = lPhys + lBio weighting it by means of action spectra which take into account how different wavelengths of optical radiation affect the human body. The organs exposed to optical radiation are the skin and Effective point of measurement the eyes. (Radiotherapy) The dose is the expectation value of the energy Exposure for the skin is measured by energy received for unit imparted per unit of mass of an infinitesimal volume centred in a area in Jm−2 (radiant exposure). Exposure for the eyes is normally specific point. The dose takes a value at each point of an irradi- expressed as energy through a unitary solid angle in Jm−2sr−1 ated medium and describes the spatial distributions of the energy (radiance). imparted. The point at which the dose is specified is indicated as The exposures are estimated taking into account the biological reference point (point of interest) at the reference depth. interaction of light according to its energy (i.e. its wavelength), by In practice the measurement of the dose is performed intro- weighting the exposure by photobiological action spectra. ducing a detector into the irradiated medium and applying the There are three key action spectra published by ICNIRP in its cavity theory. A radiation detector responds to radiation with a guidelines: signal which is basically related to the energy imparted to the detector volume by the application of several coefficients, per- • S(λ): An ultraviolet action spectrum, which is based on turbation and correction factors. To determine dose at the refer- the light-skin interactions ence point, protocols and codes of practice have been introduced • B(λ): A blue light action spectrum, which takes into ensuring that procedures and physical data sets are the same. The account the photochemical interactions between visible protocols are based on the use, as radiation detector, of an ionisa- light (primarily around the blue region) tion chamber (i.c.) traceable to a national or international primary • R(λ): A thermal action spectrum which takes into standard. The dose is determined at the reference point (at the account the heating hazards on both the eye and skin of reference depth) in a specific beam (type, quality) in the absence visible and infrared radiation. of the chamber. Effective point of measurement, Peff, is the point within The spectra are also tabulated into the AORD Directive. the chamber to which the determined dose is related. For the Related Articles: Action spectra, AORD, Maximum exposure standard calibration geometry, i.e. a radiation beam incident time, ICNIRP, Eye, Skin from one direction, Peff is shifted from the position of the cen- Further Readings: Coleman, A., F. Fedele, M. Khazova, P. tre towards the source by a distance which depends on the type Freeman and R. Sarkany. 2010. A survey of the optical hazards of beam and chamber and on the used dosimetry protocol. Effective source point 315 E FOMP Measurement electron beams with different energies will appear to have origi- media nated at different source positions. Source position is an important factor when calculating the change in output factor for extended SSD treatments. Abbreviation: SSD = Source to surface distance. P Related Article: Apparent source position Further Reading: Podgorsak, E. B. 2005. Radiation Oncology w Physics: A Handbook for Teachers and Students, International Atomic Energy Agency, Vienna, Austria. Effective SSD

(Radiotherapy) The effective SSD of an electron beam is the distance from the effective (or apparent or virtual) source point (or position) to Measurement the surface of the irradiated object. The effective SSD changes with media changing beam energy. Methods to find the effective SSD and/or the effective source point are given in the following references. The term effective SSD is also used to describe a method employed to correct for oblique beam incidence or an irregular Peff patient contour (see related articles for more details). E Abbreviation: SSD = Source to surface distance. Related Articles: Apparent source position, Effective source point, apparent focal spot, Electron oblique incidence, Oblique incidence Further Readings: Klevenhagen, S.C. 1985. Physics of Electron Beam Therapy Medical Physics Handbooks 13. Adam deff Hilger Ltd.; Podgorsak, E.B. 2003. Review of Radiation Oncology Physics: A Handbook for Teachers and Students. International Atomic Energy Agency. FIGURE E.13 Shift of the effective point of measurement. Effective voltage value (General) Effective voltage value is the voltage value that has Although the protocols are careful to point out that the recom- the same effect and gives the correct result on a power calcula- mended shift of the chamber only works for depths beyond the tion as does a DC voltage of the same value. Effective voltage is depth of maximum dose, this prescription is routinely applied equal to the square root of the mean value of the squares of the in relative photon beam dosimetry for all depths due to the lack magnitudes of an AC voltage measured at each instant over a of alternative recommendations. For plane-parallel ionisation defined period of time, usually one cycle. The effective voltage chambers Peff is usually assumed to be situated in the centre of is also known as the RMS (root mean square) voltage (see the the front surface of the air cavity. Peff is positioned at the refer- eponymous article). ence depth (Figure E.13). Related Article: RMS (root mean square) voltage An alternative to the use of the effective point of measurement of the chamber is to put the chamber centre to the reference depth EFOMP and to introduce a perturbation factor pdis that accounts for the (General) The European Federation of Organisations for Medical effect of replacing a volume of media with the detector cavity. Physics (EFOMP) was founded in 1980 as the first Regional However, in some protocols such as the IAEA TRS 398 (2000), Organisation of IOMP. Currently (2019) the Federation consists the effective point of measurement is utilised only for electron of 34 national member organisations and three affiliated national and heavy ion beams. For other beams (60Co, high energy photon organisations, representing about 10,000 physicists and engineers beams, kilovoltage x-ray beams and protons) the reference point working in the field of medical physics. of the i.c. (for plane parallel chambers on the inner surface of the Since its inauguration, the main objective of EFOMP has been window at its centre, for cylindrical chambers on the central axis to harmonise and promote the best practice of medical physics at the centre of the cavity volume) is positioned at the reference in Europe. In order to accomplish this goal, EFOMP has pre- depth where the dose is determined. sented a number of unanimously adopted policy statements, mak- ing recommendations on the appropriate general responsibilities Effective source point and roles of the medical physicist and proposing guidelines for (Radiotherapy) Electron beams appear to originate from a point education, training and accreditation programs in medical phys- in space that does not coincide with the scattering foil or the ics. The aims and purposes of EFOMP also include collaboration accelerator exit window. The term effective (or apparent or vir- with national and international organisations and encourag- tual) source point (or position) was introduced to indicate the vir- ing exchange and dissemination of professional and scientific tual location of the electron source. information. In 2006 EFOMP formed a legal body – EFOMP Effective source point is relevant to electron beams since they Company. have passed through a scattering filter. This will change the beam EFOMP includes the medical physics societies from the fol- from its well defined collimated shape to one which diverges. The lowing countries: Austria, Belgium, Bosnia and Herzegovina, degree of divergence will be energy-dependent and will mean that Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Eigenfunctions 316 e-Learning Finland, France, Germany, Georgia, Greece, Hungary, Ireland, Elastography Italy, Latvia, Lithuania, North Macedonia, Malta, Moldova, (Ultrasound) Ultrasound elastography is the use of ultrasound to Netherlands, Norway, Poland, Portugal, Romania, Russia, Serbia, produce measurements or images of tissue stiffness. In order to Slovakia, Slovenia, Spain, Sweden, Switzerland, UK. Affiliated produce elastography images, a force must be applied to produce members: Algeria, Israel, South Africa. a strain. The nature and source of the force used to produce a International Collaboration: EFOMP is part of a larger strain can take very different forms, e.g. a force applied by the international medical physics network through the International operator, an external mechanical device, physiological motion or Organization for Medical Physics (IOMP), where EFOMP is an acoustic force generated by the ultrasound probe. an affiliated regional organisation. IOMP, together with the It is possible to derive a stiffness image by comparing an International Federation of Medical and Biological Engineering image acquired before a stress is applied, with one acquired after. (IFMBE), are the constituent organisations of the International To improve performance, most implementations compare raw Union for Physical and Engineering Sciences in Medicine ultrasound data across several frames. A hard object will tend to (IUPESM). IUPESM is an international network of physical sci- move as a whole, whereas soft tissue compresses more unevenly, entists and engineers dedicated to improving healthcare and well- with tissues closer to the applied force compressing more than being worldwide, especially in developing countries. In its role those further away. These features are illustrated by Figure E.14. of an International Scientific Union, IUPESM is a member of the The simplest method of obtaining data under different stresses International Science Council (ICS, before 2019 ICSU). is for the operator to provide a force. This can be, e.g. from a slow, Hyperlinks: EFOMP: www .efomp .org; IOMP: www .iomp steady pressure or a gentle bouncing motion. However, because E .org; IFMBE: www .ifmbe .org; IUPESM: www .iupesm .org; ISC: of the operator dependence, these methods do not provide fully www.council.science quantitative data of tissue stiffness. There are several alternative techniques. Sonoelastography Eigenfunctions uses an externally applied low-frequency vibration as the stress, (Nuclear Medicine) For the linear operator L, an eigenfunction and makes use of Doppler techniques to derive the strain. defined on a function space is every nonzero function, f, which Vibroacoustography uses an oscillating acoustic force to apply returns the original value of the function multiplied by a discrete the stress. A more recent development is that of acoustic radiation factor λ also called eigenfunction: force imaging (ARFI). This technique produces a push-pulse of high intensity ultrasound to displace the tissue, and then lower intensity pulses to image. This technique may enable quantita- Lf = lf (E.2) tive measurements to be made. Several suppliers are working on bringing ARFI-based systems to market, but at the time of writ- ing none are yet available. Eigenvalues (Nuclear Medicine) Eigenvalues λ are a set of scalars for a specific e-Learning linear system of equations. (General) The term e-learning gained popularity in 1998 through Further Reading: Weisstein, E. W. Eigenvalue, From the paper by A. Morri in Connected Planet (November 1997) ‘A MathWorld – A Wolfram Web Resource. http: / /mat hworl d .wol bright future for distance learning: One Touch/Hughes alliance fram. com /E igenv alue. html (accessed 31 July 2012). promotes interactive “e-learning” service’. Approximately around this time, many people started using the term e-learning. There Elastic scattering are various definitions of e-learning. If one selects some of these, (Radiation Protection) Also known as coherent scattering, or they would vary from shorter ones, ‘e-learning – learning that is Rayleigh scattering. aided by information and communication technologies’ or ‘learn- When the distance between a charged particle and an atom ing conducted via electronic media, typically on the internet’ to is less than the diameter of the atom, the particle experiences a longer ones such as, ‘e-learning is a method that makes educa- Coulomb-force interaction with the nucleus. The interaction may tional content available on electronic media (CD-ROM, internet, take the form of either elastic scattering or radiation energy loss. These interactions are only important for light charged particles such as electrons and positrons. Most of these events are elastic scattering, in which the elec- Object with a harder Object with a softer tron is deflected with no significant transfer of energy to the inclusion inclusion medium. This is the cause of the tortuous path electrons follow through a medium. The likelihood of elastic scattering varies with Z2. Most elastic scattering events at diagnostic x-ray and low Uncompressed objects gamma ray energies are defined as Rayleigh scattering. The 2%–3% of events that are inelastic are termed radia- tion interactions. The incident light charged particle is decel- erated and deflected by the electric field of the nucleus and emits a significant proportion (up to 100%) of its energy as a Bremsstrahlung x-ray. The cross section for radiation events var- ies with the square of the absorber atomic number Z2, as well as Compressed objects EK and 1/m2 where EK and m are, respectively, kinetic energy and mass of the particle. Related Articles: Coherent scattering, Rayleigh scattering, Inelastic scattering FIGURE E.14 Different behaviour of hard and soft inclusions. Electric arc 317 Electric dipole intranet, extranet, interactive TV, etc.) encompassing content as Usually electric arc is considered to be the strong current of well as educational tools and applications’, etc. a continuous discharge, while momentary electric discharges are Medical physics is one of the first professions to embrace considered electric sparks. e-learning. The EMERALD project (1995–1998) was devel- Related Article: Arcing of x-ray tube oped and introduced before the existence of the term e-learning. It developed one of the first educational image databases and Electric current related e-learning materials in the world (it was followed by the (General) Electric current is the flow of electric charge carriers. EMERALD II, EMIT and EMITEL e-Encyclopaedia projects). In The electric charge carriers may be either electrons or ions. In parallel with these projects a number of other e-learning projects a wire, electric current is a flow of electrons that have been dis- were developed and successfully introduced in medical physics lodged from atoms and is a measure of the quantity of electrical practice. To mention a few: charge passing any point of the wire per unit time. In gases and • NICER – Electronic Encyclopaedia of Medical Imaging liquids the electric current is flow of positive ions in one direction (mainly medical) together with a flow of negative ions in the opposite direction. • Medcyclopaedia (mainly medical) Conventionally, the direction of electric current is that of the flow • Sprawls Resources of the positive ions. In alternating current (AC) the motion of the • IAEA Training Course on Radiation Protection in charges is periodically reversed. The SI unit of electric current Diagnostic and Interventional Radiology intensity is the ampere, a flow of 1 C of charge per second, or • Demystifying Biomedical Signals 6.24 × 1018 electrons per second. • A web-based course on Medical Physics for E Schoolteachers • Virtual Library of the American Association of Electric dipole Physicists in Medicine (AAPM) (General) The electric dipole is the combination of two equal • IAEA RPOP website point charges of opposite sign q separated by a distance a. The • X-ray equipment simulator dipole moment is defined as • PRISM and VERT projects • IAEA SAFRON project p = qa • EUTEMPE and EMETRAP projects, etc. e-learning allows better pedagogical effectiveness and man- The electrostatic potential from a dipole at a point P is given by agement of learning. It is very suitable for dynamic professions. The first international conference on e-learning in medical phys- 1 æ q q ö 1 q (r2 - r1 ) ics was organised in 2003 at ICTP, Trieste (part of project EMIT). V = ç - è r r ÷ = 4pe 0 1 2 ø 4pe0 r1r2 The rapid

expansion of e-learning in medical physics was one of the triggers for the establishment of the IOMP Journal Medical Physics International – a free online journal (www .mpijournal where r1 and r2 are, respectively, the distance of the positive and .org) which provides a free educational platform for various ini- negative charge from P (Figure E.15). tiatives, including e-learning projects. When a ≪ r the electrostatic potential V becomes: A number of special software applications allow building e-learning courses and modules. Moodle is one of the three most 1 qacosq 1 pcosq popular such applications. V = 2 = 4pe0 r 4p 0 r2 e Related Articles: EMERALD, EMERALD II, EMIT, EMITEL, RPOP website, Sprawls resources, Moodle In spherical polar coordinates the components of the electric field Further Readings: Tabakov, S. 2005. Special issue on of a dipole are given by: e-Learning in Medical Engineering and Physics. J. Med. Eng. Phys. 27(7); Tabakov, S. and V. Tabakova. 2015. The Pioneering of e-Learning in Medical Physics, Valonius Press, London, available free at: www .emerald2 .eu /mep _15 .html; Tabakova, V. P 2020. e-Learning in Medical Physics and Engineering: Building Educational Modules with Moodle, CRC Press. Electric arc (Diagnostic Radiology) Electric arc (or voltaic arc) is a phenom- r2 enon of electrical current flowing through normally non-conduc- r tive medium, driven by high electric voltage. r The gas discharge which occurs in some (old) x-ray tubes is a 1 phenomenon similar to short electric arc (or long electric spark). It can occur at situations when gas ions are allowed inside the tube envelope (either due to micro cracks in the glass envelope, or r2 – r1 gas extraction from the electrodes due to the high vacuum). These ions, being in the field of high voltage between the anode and the θ cathode, create a large spark (or cluster of sparks) which results in – q O + q very high anode temperature and anode current (plus bright light) a and could destroy the x-ray tube. X-ray generators often have fast fuses and spark distinguishers to prevent such problems. FIGURE E.15 The electric dipole. Electric field 318 E lectromagnet 2pcosq where q is a small charge assumed positive placed at that point. Er = The direction of the electric field is the same as the direction of the 4pe 3 0r force it would exert on a positively-charged particle and opposite to the direction of the force on a negatively-charged particle. The psinq Eq = SI unit of electric field intensity is the Newton per coulomb (N/C). 4pe r3 0 An equivalent unit for the electric field intensity is the volt per where meter (V/m). Electric fields follow the superposition principle and E therefore, at any point, the total electric field created by more than r is the radial component Eθ is the transverse component of the electric field (Figure E.16) one charge is equal to the vector sum of the respective electric fields that each charge would create in the absence of the others. In Figure E.17 the lines of force of the electric dipole are shown. Electric power (General) The energy dissipated by an electric current per unit of Electric field time. It is calculated by multiplying the current consumption by (General) An electric charge or a distribution of charges will the applied voltage. The SI unit of power is the Watt. cause a force F to act on some other charge located in the sur- Related Article: Watt rounding space. The electrical field intensity denoted as E at a Electrical charge E point in the space is defined as (General) See Charge F E = q Electrical interlock (Radiotherapy) See Interlock; Interlocking device Electrical resistance (General) See Resistance, electrical ε Electricity, static (General) See Static electricity ε ε r 0 P Electrocardiographic triggering (General) See Cardiac gating r Electrode Line of force (General) An electrical terminal is used to make contact between metallic and non-metallic parts of an electric circuit. It is most u0 θ often used to pass a current through an electrolyte, gas, vacuum or u semiconductor. Electrodes are usually in the form of rod, plate or r P Z a wire. Metal electrodes may be made of a copper, lead, platinum, silver, or zinc. Non-metal electrodes are commonly made of car- bon. The electrode through which current passes from the metal- FIGURE E.16 The radial and transverse component of the electric lic to the non-metallic conductor is called the anode, and that dipole. through which current passes from the non-metallic to the metal- lic conductor, the cathode. Typical electrode applications are elec- trochemical, electrolytic and voltaic cells, electrolytic capacitors, rechargeable batteries, vacuum tubes and semiconductors. Electrolytic capacitor (General) See Capacitor Electromagnet (Magnetic Resonance) An electromagnet produces a magnetic field by an electric current through the conductor geometry. The P simplest design is the solenoid (Figure E.18) producing a homo- geneous field within the ‘coil’ and a dipole field at a distance. By placing ferromagnetic material (commonly soft iron) inside the coil the outside magnetic field is enhanced. The ‘toroid’ design is a closed loop solenoid to produce a strong homogeneous field in the gap between the terminal shoes. The design of superconducting MR-scanners or spectrometers is commonly solenoidal. The open design of resistive MR-scanners usually features two coils in either a horizontal ‘double doughnut’ FIGURE E.17 Lines of force of the electric dipole. (comparable to a solenoid) or a vertical C-shape configuration Electromagnetic energy spectrum 319 Electromagnetic field (comparable to a toroid) (Figure E.19). The former design features air cores and static fields of maximum flux density B0 = 0.2 T, the I latter 0.6 T with an iron core. Related Articles: Magnet, Permanent magnet, Resistive mag- net, Superconductive magnets N Electromagnetic energy spectrum (General) The electromagnetic spectrum is a collection of dif- ferent electromagnetic wave types. Electromagnetic waves are energy waves that are propagated by paired, transversely oscil- lating electric and magnetic fields and can be classified as one of the following types – radiowaves, microwaves, infrared, visible, FIGURE E.18 A conductor is winded around a core of a para- or ferromag- ultraviolet, x-rays or gamma rays. netic material. When a current, I, is passed through the conductor a magnetic These wave types are defined by different wavelengths, fre- field is generated. The polarity depends on the direction of the current. quencies or energies (Figure E.20). The spectrum can be divided into two unequal parts – ionising radiation and non-ionising radi- ation. Most types of wave in the spectrum are employed across the different areas of medical physics and clinical engineering. There are no clear boundaries between the different wave types E and they tend to overlap with their neighbours. The different frequencies and wavelengths dictate how the waves will interact I with matter. Related Articles: Radiation, Ionising radiation, Non-ionising radiation, Ultraviolet radiation, Infrared radiation, Microwaves B0 Electromagnetic field (Magnetic Resonance) Electromagnetic fields are produced by electrically charged particles and can be described as either a wave or a particle with zero mass travelling through empty space with the speed of light. As an example, an antenna functions by I oscillations of charged particles along an electrically conductive structure. This causes the emission of an electromagnetic wave. The description of the electromagnetic fields as waves is then suit- able and generally applies at the lower range of frequencies (lon- FIGURE E.19 An MRI- scanner with an electro-magnet is built by two ger wavelength), e.g. for radiowaves and microwaves. Although magnets that create the static magnetic field, B0, in- between. When the both descriptions would be equally valid, at higher frequencies current, I, is switched on, the magnetic field is created. (e.g. x-rays and gamma rays) a more suitable model will be to (Wavelength–in metres) 103 102 10 1 10–1 10–2 10–3 10–4 10–5 10–6 10–7 10–8 10–9 10–10 10–11 10–12 Radiowaves Gamma-rays Microwaves X-rays Infrared Ultraviolet (Frequency–in Hertz) 106 107 108 109 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 (Energy–in eV) 10–9 10–8 10–7 10–6 10–5 10–4 10–3 10–2 10–1 1 10 102 103 104 105 106 FIGURE E.20 A simple diagram illustrating the arrangement of wave types within the electromagnetic spectrum, with approximate values for fre- quency, wavelength and energy. Visible Electromagnetic hyperthermia 320 Electron describe electromagnetic fields as particles with a specific energy a problem. One important factor is the rate of cooling which is and zero mass, denoted photons. related largely to blood flow. Electromagnetic fields can also be described separately as an Related Article: High-intensity focused ultrasound (HIFU) electrical field and a magnetic field. Stationary charged particles Further Reading: Johns, H. E. and J. R. Cunningham. can then be considered the source of the electrical field and mov- 1983. The Physics of Radiology, 4th edn., Charles C. Thomas, ing charged particles (current) can be considered the source of the Springfield, IL, pp. 706–707. magnetic field. The electromagnetic fields can be described mathematically Electromagnetic spectrum by Maxwell’s equations. For details about these very fundamen- (General) See Electromagnetic energy spectrum tal relationships, please refer to any suitable educational physics textbook. Electrometer In MRI three types of electromagnetic fields are encountered: (Radiotherapy) The charge produced in an ionisation chamber (i.c.) under typical irradiation condition is small. An exposure 1. The static magnetic field, which, e.g. interacts with rate of 2.58 × 10−4 C/kg min produces in 1 cm3 volume containing moving charged particles and objects. approximately 1.29 × 10−6 kg air at standard conditions an ionisa- 2. The switching magnetic gradient fields (in the order of tion current of 5 × 10−12 A. Depending on the ionisation cham- 1 kHz) which interact with, e.g. the patient’s body and ber volume and the exposure rate the currents range from about cause induced currents, potentially leading to nerve 10−14 to 10−10 A below the typical detection limit of a galvanom- E stimulation. eter which is of the order of 10−9 A. These small currents may 3. The radiofrequency field (42.6 MHz/T) which inter- be measured connecting the i.c. in series with a sensitive current acts with the nuclei at resonance but which also causes device or connecting the i.c. to a charge measuring device. In the heating through induced current and interaction with former case the current is measured during the irradiation while dipoles in the body. in the latter the charge accumulated during an interval of time is measured. The charge measuring device is called electrometer. In order to limit adverse effects of electromagnetic radiation in An electrometer used in conjunction with an ionisation chamber MRI legal limits apply. is a high gain, negative feedback, operational amplifier with a Related Articles: Radiofrequency, Radiowaves, Eddy cur- standard resistor or a standard capacitor in the feedback path to rents, Gradient field measure the chamber current or charge collected over a fixed time interval. In Figure E.21 the circuit diagram of an electrometer Electromagnetic hyperthermia used in integrating mode is shown. (Radiotherapy) Hyperthermia involves the use of heat, usually in Several requirements are needed for an acceptable perfor- the temperature range 37°C–50°C, as a technique for the treat- mance of the electrometer: ment of cancer patients. Studies in tissue cultures have shown that animal and human tumour cells are sensitive to temperature • Zero stability with fast warm up and long-term drift changes of a few degrees and that tumours can regress follow- • Linearity ing heating. In addition, heating can raise the radio-sensitivity of • Noise level tumour cells and reduce their ability to repair radiation damage. • Input bias and offset requirements Because of this hyperthermia has been employed with radiother- • Speed of response and slew rate apy, and more recently with chemotherapy, for the treatment of • Ruggedness cancer. Ultrasound as well as electromagnetic radiation (electromag- Electron netic hyperthermia) such as radio-frequency electric fields and (General) The electron is a negatively charged subatomic par- microwaves have all been used for hyperthermia. It is impor- ticle (Table E.2). It has been discovered in 1897 by Joseph John tant to note that temperature has to be carefully controlled since Thomson. In the classical model of the atom, electrons exist in changes of as little as 0.5°C can lead to significant changes in orbits around the positively charged nucleus of the atom. The biological effect. number of electrons in an uncharged (neutral)

atom equals the There are three main types of hyperthermia: 1. Local hyperthermia heats a very small area, usually the tumour itself. Depending on the location of the tumour, the heat may be applied to the surface of the body (superficial hyperthermia), inside normal body cavities, or deep in tissue through the use of needles or probes (interstitial hyperthermia). 2. Regional hyperthermia heats a larger part of the body, High R C such as an entire organ or limb. Usually, the goal is to voltage Measurement weaken cancer cells so that they are more likely to be signal killed by radiation and chemotherapeutic medications. 3. Whole-body hyperthermia heats the entire body. It is Ground typically used to treat metastatic cancer. Because of their varying thermal characteristics, achieving and FIGURE E.21 Circuit diagram of an electrometer used in charge mea- maintaining optimal temperature in different tissues remains surement mode. Electron angular distribution 321 Electron angular scattering power number of protons in the nucleus, given by the atomic number of More correctly, in the quantum model, electrons exist in the atom. zones of probability around the nucleus called orbitals rather Electrons occupy shells in the atom defined as K, L, M, etc. than precise orbits. The K, L and M correspond to principal shells moving out from the nucleus of the atom. Electrons closest quantum numbers of 1, 2, 3. The full set of quantum numbers to the nucleus have the highest electronic binding energies. is given by the principal, orbital, magnetic and spin quantum numbers. Under the Pauli exclusion principle no two electrons in an atom can have the same set of quantum numbers. This principle underlies the electronic structure of the elements and TABLE E.2 determines how many electrons can occupy a given electronic Properties of the Electron ‘shell’. Charge −1.602 × 10−19 C Rest mass 0.11 × 10−31 kg Electron angular distribution Rest mass energy 0.511 MeV (General) The recoil electron angular distribution in Compton effect is a plot of the probability of scattering versus angle for an electron involved in a Coulomb interaction event. The distribu- tion will be a function of the photon energy and the scattering y medium. The following plot shows an approximate angular distri- bution for emission of Compton electrons generated by a 1 MeV pv΄ cos θ incident photon (Figures E.22 and E.23). E Recoil pc sin φ electron The photon scattering angle θ and the recoil electron angle ϕ P υ c = mc are related through the following relationship: υ 2 1 = c Incident q photon φ cotf = (1+ e) tan 2 θ x pv = hν where ɛ is the incident photon energy hν normalised to the rest c pc cos φ mass energy of the electron mec2 = 0.511 MeV. Further Reading: Podgorsak, E. 2006. Radiation Physics for pv΄ sin θ Scattered Medical Physicists, Springer, Berlin, Germany. photon pv΄= hν΄ c Electron angular scattering power (General) The electron angular scattering power is a measure of FIGURE E.22 Schematic diagram of Compton effect. (Drawing cour- the electron angular scattering distribution. The ICRU defines tesy of E. Podgorsak.) the mass angular scattering power T/ρ as the rate of change Distribution of scattered photons Incident photons Distribution of recoil electrons FIGURE E.23 Recoil electron angular distribution plotted in polar coordinate system. (Adapted from Springer Science+Business Media: Radiation Physics for Medical Physicists, 2006, Podgorsak, E.) The length of a line from the origin to the distribution curve is proportional to the probability of the electron scattering at that angle. Electron applicator 322 Electron backscatter factor of the mean square scattering angle dq2 per unit mass thickness with the gantry in motion in order to treat an extended area, e.g. ρds traversed: the chest wall. This is called electron arc therapy. Electron arc aperture is the special beam applicator or tray (insertion) which is inserted into the beam limiting system to T dq2 = enable electron arc therapy. r rds Related Articles: Electron beam, Arc therapy, Curvature correction and provides the following expression for calculating T/ρ: Electron arc therapy 2 T N é 1+ t ù ì 2 -1 ï æ q ö é q2 ù ü (Radiotherapy) Electron arc therapy is a form of electron beam = 4p A r2 e Z (Z +1) ê ú ç1+ max r ëê t( ÷ -1+ ï 2 + t) íln 2 ê1+ max ú ý A radiotherapy delivered by linear accelerator. The treatment tech- ûú îï è qmin ø ë q2 min û þï nique is designed for treatment of large area underneath and across the chest wall or a superficial target volume involving where a large area on or near the body surface. A specially designed electron beam collimator shutter instead of an electron treatment 2A-1/3 applicator is used for treatment delivery. The collimator shutter qmax = ab(t +1) produces an elongated but narrow strip of radiation field of about 2 cm wide and a length that matches the length of the treatment E is the cut-off angle due to the finite size of the nucleus given by target. Unlike photon beam arc therapy where the treatment iso- the ratio of the reduced de Broglie wavelength of the electron to centre is located at or near the centre of the target volume, the the nuclear radius, and isocentre of electron arc therapy is located relatively far behind the target volume, i.e. a small focus to skin distance is used. The aZ1/3 gantry rotates slowly during treatment delivery, and this delivers qmin = 1.13 b(t +1) a narrow strip of uniform electron beam around the body surface beneath which the target volume is located. The gantry speed, gantry start and stop angles, dose per gantry angle rotation, iso- is the screening angle due to the screening of the nucleus by the centre location, and field length are chosen such that the treatment atomic orbital electrons with can deliver a uniform dose to the target volume. The treatment can α, the fine structure constant (1/137) have large penumbra. To better protect the critical normal tissues β, velocity of electron normalised to speed of light in vacuum near the field edges, lead sheet of appropriate thickness and with τ, ratio of electron kinetic energy to its rest energy appropriate cut-out is sometimes placed on the patient to improve A, nucleon number the sharpness of the shielding. Another potential problem with the Z, atomic number treatment is the possibility of delivering a high x-ray dose to the isocentre area. The x-ray dose is due to bremsstrahlung radiation Further Readings: ICRU (International Commission on that is focused at the isocentre during electron arc treatment. Radiation Units and Measurements). 1984. Radiation dosimetry: Electron beams with energies between 1 and 50 MeV. ICRU Electron backscatter factor Report 35, Bethesda, MD; Podgorsak, E. 2006. Radiation Physics (Radiotherapy) The presence of any inhomogeneity (e.g. lung, for Medical Physicists, Springer, Berlin, Germany. bone) within the path of a beam will significantly modify the elec- tron dose distribution due to different absorption properties and Electron applicator effects on the electron scattering. Therefore, when an inhomo- (Radiotherapy) In order to provide a usable electron treatment geneity of high atomic number (relative to water) is in the beam, beam, it is necessary to attach an electron applicator (sometimes electrons being backscattered from the inhomogeneity may result called cone) to the head of the linear accelerator. These applica- in an increased dose at the interface between the tissue and the tors typically come in a range of set field sizes (e.g. 6 × 6, 10 × 10, inhomogeneity. 15 × 15, 20 × 20 and 25 × 25 cm2). Examples where internal shielding is used to protect normal The applicator is needed because the penumbra produced structures beyond the target include treatments of the lip, buccal without it would be clinically unacceptable. This is due to the fact mucosa or ear. Care must be taken as the electrons backscattered that while some beam shaping is provided by the secondary col- from the shielding (usually lead) can deliver a high dose to the limators in the head of the linear accelerator there is a significant healthy tissue in contact with the shielding. Therefore it is com- amount of scatter both within the linear accelerator and in the air mon to coat the shielding in a few millimetres of wax which will between the accelerator and the patient. Therefore the applicator absorb the low energy scattered electrons. collimates the beam and defines it typically at a distance of 5 cm To quantify the change in dose due to the backscattering a from the patient. Some applicators are also used to provide addi- property known as the electron backscatter factor (EBF) may be tional electron scatter thus improving the flatness of the beam. determined. The EBF is defined as the ratio of the dose at the If fields other than the sizes produced by the applicator set are interface surface with and without the inhomogeneity present. required, it is common to create an appropriately shaped alloy The EBF has been found to increase with increasing atomic num- that can be inserted into the end of the applicator. ber and decrease with increasing beam energy. Related Article: Collimation Abbreviation: EBF = Electron backscatter factor. Related Articles: Backscatter, Backscatter factor, Bone soft Electron arc aperture tissue interface, Inhomogeneity correction factor (Radiotherapy) Electron treatment can be delivered by static Further Readings: Klevenhagen, S. C. 1985. Physics of fields with no movement of the gantry. It can also be delivered Electron Beam Therapy Medical Physics Handbooks 13, Adam Electron beam 323 Electron capture (EC) decay Hilger Ltd, Bristol, UK; Klevenhagen, S. C., G. D. Lambert and x-rays produced at the target are collimated to a fan beam 47 cm A. Arbabi. 1982. Backscattering in electron beam therapy for diameter at the isocentre. energies between 3 and 35 MeV. Phys. Med. Biol. 27(3):363–373. The detector array consists of two solid-state arcs of 216°, opposite the target rings. The scanner can therefore acquire two Electron beam simultaneous slices when one target ring is used, or eight slices, if (Radiotherapy) Linear accelerators are used to provide clinical all four targets are used in sequence. beams of electrons which usually have nominal energies in the Its high temporal resolution made this scanner particularly range 4–20 MeV. Electrons are produced by thermionic emission suitable for cardiac applications. However, because of limitations and injected into a waveguide structure operating in conjunc- in terms of power output, slice thickness and volume coverage, it tion with microwave radiation at a frequency of 3000 MHz. The has been largely superseded by the latest models of multislice CT microwave energy required to accelerate the electrons is delivered scanners. to the accelerating structure in the form of short duration pulses Related Articles: Imatron, Multislice CT scanner at a repetition rate in the range 50–300 Hz. The electron beam, with a typical diameter in the range 4–6 mm, has a maximum Electron beam dosimetry energy and a narrow beam energy spread (a desirable feature), (Radiotherapy) The absorbed dose delivered by an electron which depends on the accelerator design (full width half maxi- beam can be measured by several methods but it is often deter- mum 40 keV for a microtron and 50 keV for a linear accelerator). mined by ionisation chamber measurements. The theoretical The beam then passes through the exit window and enters the foundations of absorbed dose dosimetry in an electron beam are beam handling and monitoring system of the accelerator. Linacs given by the Bragg–Gray cavity theory. Electron beam dosim- E are equipped with a scattering and collimation system to achieve etry is more complex than photon beam dosimetry because field flatness and symmetry for a 40 × 40 cm2 field at one meter. of the characteristics of the electron interaction with matter. The most common scattering technique utilises a dual set of foils Electrons are charged particles, and, therefore, the electron where the second foil is positioned at some distance from the first. energy spectrum varies with depth in a material because of their The attractive feature of electron beams used in radiotherapy interaction. In addition the polarisation density effect becomes relates to the fact that the absorbed dose decreases sharply beyond important for high energy electrons and consequently

the dose a certain depth in the patient and this depth varies with beam conversion factor for an ionisation chamber in a medium var- energy. Electron beams can therefore be used to treat superficial ies with depth. Knowledge of the electron energy spectrum is cancer at depth up to about 5 cm and are generally as a single field crucial for absorbed dose determination since the fundamental irradiation. physical parameters, such as the stopping power of a material, are a function of energy. Many international codes of practice Electron beam CT and protocols have been published for the dosimetry of electron (Diagnostic Radiology) The electron beam CT (EBCT) scanner is beams under reference conditions. It is essential, because of the a novel concept of CT scanner design with no mechanical motion. dosimetric considerations, that a strict adherence to the used It was originally marketed as the Imatron scanner, but currently is protocol is kept without mixing together recommendations from marketed by GE Medical Systems as the eSpeed. different protocols. In place of a rotating gantry with x-ray tube, the EBCT employs an electron beam which is accelerated and electromagnetically Electron capture steered around an anode, consisting of four tungsten rings form- (Radiation Protection) Electron capture, sometimes known as ing an arc of 210° around the patient (Figure E.24). One sweep of K-capture, is where an electron is captured during radioactive the arc results in a partial scan within a time of 33–100 ms. The decay of a radionuclide from one of the electron orbitals, most often the K-shell, of a radionuclide and combines with a proton in the nucleus with the formation of a neutron and a neutrino: Data acquisition system P + e ® n + n Detector ring Related Articles: Electron capture (EC) decay, Neutron, Electron gun Target-ring Neutrino, Radioactive decay, Radionuclide Electron capture (EC) decay Couch (Radiation Protection) Electron capture, sometimes known as K-capture, is a type of radioactive decay where an electron is cap- tured from one of the electron orbitals, most often the K-shell, of a radionuclide and combines with a proton in the nucleus with the formation of a neutron and a neutrino: P + e ® n + n Electron beam The overall change in the nucleus is exactly the same as in posi- Internal cooling system tron decay. It is a different mechanism from positron decay (beta plus decay) in that no particle, apart from a neutrino, is emitted FIGURE E.24 Schematic diagram of basic components of an electron from the nucleus but the net effect in the nucleus is the same – one beam CT scanner. less proton and one more neutron. Electron contamination 324 Electron dual scattering foils 51 24Cr 27.8 d TABLE E.3 EC1 Electron Densities for Some Tissues EC2 Electron Electron 320 keV Physical Density Density γ Density per Relative to Material (g/cm3) cm3 × 1023 Water (RED) 0 51 23V Water 1.00 3.340 1.000 Lung (inhale) 0.20 0.634 0.190 Lung (exhale) 0.50 1.632 0.489 FIGURE E.25 The decay of 51Cr by electron capture. Adipose 0.96 3.170 0.949 Breast (50/50) 0.99 3.261 0.976 An example is the decay of chromium-51 to vanadium-51. Muscle 1.06 3.483 1.043 Ninety-one per cent of 51Cr atoms decay directly to the ground Liver 1.07 3.516 1.052 state of 51V and 9% via the 320 keV level with resulting emission Trabecular bone 1.16 3.730 1.117 of a 320 keV gamma ray. The K-shell electrons are involved in Dense bone (800 mg/cc) 1.61 5.052 1.512 E 90% of captures, the remainder from the L-shell (Figure E.25). Dense bone (1000 mg/cc) 1.66 5.243 1.570 Related Articles: Beta decay, Electron, Neutron, Neutrino, Dense bone (1250 mg/cc) 1.83 5.718 1.712 Positron, Positron decay, Radioactive decay, Radionuclide Dense bone (1500 mg/cc) 2.00 6.209 1.859 Dense bone (1750 mg/cc) 2.17 6.698 2.005 Electron contamination Titanium 4.51 12.475 3.735 (Radiotherapy) The photon beam exiting the Linear Accelerator Treatment head has some low energy electron contamination that arises from photon interactions (predominantly Compton) within the treatment head. These secondary electrons contribute 2.5 to the dose received by the patient, particularly by increasing the surface dose. The range of contamination electrons for 21 MV photon beam is about 6 cm. Therefore 10 cm is recommended as 2 reference depth in some dosimetric protocols. Any scattering material placed into the beam increases the 1.5 electron contamination and consequently reduces skin sparing. A noticeable effect of electron contamination is a reduction of build- up region occurring with increasing field size, due to the greater 1 area of scattering material visible to the beam. To reduce electron contamination, beam modifiers and acces- 0.5 sory trays are generally placed as far away from the patient as possible. Related Articles: Build-up region, Build-up dose 0 Further Reading: Dutreix, A., B. E. Bjarngard, A. Bridier, B. –1000 –500 0 500 1000 1500 2000 2500 Mijnheer, J. E. Shaw and H. Svensson. Monitor Unit Calculation CT number for High Energy Photon Beams, Physics for clinical radiother- apy, Booklet No. 3. ESTRO (European Society for Therapeutic FIGURE E.26 Relative electron density versus CT numbers. Radiology and Oncology), 1997, Brussels, Belgium. Electron density CT scanner should be provided using phantom with inhomoge- (Radiotherapy) The electron density is the number of electrons neities with known RED. In Figure E.26 a typical CT calibration per volume unit. In Table E.3 electron densities for some tissues curve is given. Usually calibration curves are bilinear. are reported. Abbreviations: CT = Computed tomography, HU = Hounsfield The treatment planning system needs the physical and geo- unit and RED = Relative electron density. metrical characteristic of the patient to calculate the dose distri- Related Articles: CT number, Hounsfield (H) scale, Hounsfield bution with a specific radiation beam arrangement. The treatment number, Treatment planning system planning systems utilise a 3D matrix which contains electron densities of every voxel within the area of interest to incorpo- Electron dual scattering foils rate the patient heterogeneity into the dose calculation results. (Radiotherapy) The electron beam that exits the waveguide has Information about electron densities is obtained from CT num- a thin pencil beam shape, which would produce a highly local- bers, via the Hounsfield units. Published or empirically deter- ised dose without alteration. To create a clinically useful electron mined conversion tables are used for the conversion from CT beam, two scattering foils of high atomic number are used within numbers to electron densities. A formula is recommended as a the treatment head of the linear accelerator, which firstly widen default for use in the treatment planning systems in circumstances the beam, and secondly flatten it. The scattering foils are together where no data are available for a particular scanner. To achieve with the flattening filters mounted on a rotating carousel or slid- the maximum accuracy, particular CT calibration curve for each ing drawer for ease of mechanical positioning into the beam, as Relative electron density Electron field effective width 325 Electron gun Surface A A Standard measurement depth (SMD) 100% Base depth 90% B 80% (a) 50% Plane at SMD Plane at base depth Geometric (a) (b) field edge FIGURE E.27 The effect of the first (a) and second (b) scattering foil on E the beam profile. 100% 90% required. The dual foil system improves the single foil system because equally uniform beams can be produced with a con- siderably smaller total foil thickness, which reduces the energy spread. However these scattering foils will contribute to unwanted 80% Bremsstrahlung radiation, which must be accounted for by care- 90% 80% ful collimation of the beam (Figure E.27). 50% (b) Electron field effective width FIGURE E.28 The electron field effective width. (a) Beam profile in a (Radiotherapy) The effective width of an electron beam field is plan parallel to the beam direction containing the max dose. (b) Beam the distance between the points of 50% dose at the radiation beam cross section at two different depth, left half at SMD, right half at base axis along each of the major axes of the field. The distance is depth. obtained by measuring the profiles of relative doses in a water phantom in a plane parallel to the surface of the water phantom and perpendicular to the central axis. The width of the electron beam is projected by the light field of the linear accelerator. In Control Focusing Accelerating Figure E.28 isodose profiles of an electron beam are shown. In grid anode anode Figure E.28a, the profile is on a plane parallel to the incident beam direction containing the depth of maximum dose, the stan- Cathode Focussed electron dard measurement depth (SMD). The base depth is on a plane beam parallel to the SMD plane containing the 90% point on the beam central axis. In Figure E.28b, the beam cross section is shown at two different depths: left half at SMD and right half at base depth. Related Article: Field size Filament Electron Electron fluence gun (Radiotherapy) The electron fluence Φ is the quotient dN by da, where dN is the number of electrons that enter an imaginary sphere of cross sectional area da, i.e. FIGURE E.29 Block diagram of cathode ray tube. dN F = da Electrons boil off the heated cathode filament through thermionic emission and are swept along the potential difference between the The fluence is usually expressed in units m−2 or cm−2. cathode and anodes to form a stream of electrons. The control grid voltage is used to control the beam intensity: when the grid Electron gun is positive with respect to the filament, electrons are accelerated (General) In a cathode ray tube (CRT), an electron gun is used toward the screen; when the grid is negative with respect to the to produce a focused stream of electrons, which strike the screen filament, no electron current is flowing toward the screen. The phosphor (Figure E.29). accelerating anode is at a positive potential relative to the cathode The components of the electron gun are the cathode and of several thousand volts. The focusing anode ensures the electron filament, control grid, focusing anode and accelerating anode. beam is brought to a focus on the CRT screen. Electron hole pair 326 Electron off axis factor In a colour CRT, a separate electron gun is allocated to pro- 100 duce red, green and blue colours on the screen. A mask is used to 90 ensure only electrons from the correct gun excite phosphor dots of 80 the corresponding colour on the CRT screen. 70 60 Electron hole pair 50 (General) In an energy level description of conduction, the con- 40 Rmax duction band represents the electron energy levels where con- 30 duction can occur. The valence band for a material represents 20 the upper electron energy states occupied at absolute zero. In a 10 metal, the conduction and valence bands overlap, and conduction 0 can take place freely. In a semiconductor, there is an energy gap 0 10 20 30 40 50 between the valence and conduction bands. All possible electron Depth (mm) states are occupied at the top of the valence band in a semiconduc- tor. Conduction cannot take place within the valence band as all FIGURE E.32 An illustration of the electron maximum range. states are occupied (Figure E.30). For conduction to take place in a semiconductor, an electron in the valence band must absorb sufficient energy to be promoted E are incident upon. It is defined as the point at which the central to the conduction band. Promotion of an electron to the conduc- axis depth dose curve meets the Bremsstrahlung x-ray contami- tion band leaves behind an unoccupied state in the valence band nation (see Figure E.32). This is not necessarily a well-defined called a ‘hole’. As other electrons in the valence band can now measurement point, and the practical range tends to be used more move to occupy this state, in turn leaving behind a hole, holes can commonly. be thought of as charge carriers in a semiconductor. Promotion of Abbreviation: PDD = Percentage depth dose. an electron to the conduction band by, e.g., absorption of radia- Related Articles: Electron ranges, Electron practical range, tion generates an ‘electron hole pair’. Electron hole recombination Electron therapeutic range occurs where a free electron falls from the conduction band and fills a hole in the valence band (Figure E.31). Electron Monte

Carlo (Radiotherapy) For modelling the transport of electrons, for Electron maximum range example for radiotherapy treatment planning purposes, Monte (Radiotherapy) The maximum range (Rmax) is defined as the max- Carlo simulations provide an accurate solution to the underlying imum depth to which electrons can penetrate the material they transport equation, based on statistical sampling of the underlying physical processes. See Monte Carlo method. Conduction band Electron oblique incidence (Radiotherapy) Typically radiotherapy beam data (e.g. depth dose Energy gap Energy etc.) are measured with the beam incident perpendicular to the level surface of the phantom (flat and uniform). However, this situation occurs rarely. Hence these data cannot be used for dose distribu- Valence band tion calculations directly. For small angles (of typically less than 20°) the isodose lines tend to simply follow the angle of the surface contour. For angles greater than 20° there are increasingly significant changes to the FIGURE E.30 In a semiconductor all possible electron states in the depth dose characteristics – the depth of maximum dose decreases valence band are occupied and an electron gap exists between the valence and conduction bands. with increasing angle, the dose at shallow depths increases, and the dose at greater depths is reduced relative to the standard nor- mal incidence data. At large angles of incidence the dose at the maximum depth increases significantly. Conduction The most common method for understanding this problem is band to use a pencil beam algorithm for dose calculations. Related Article: Oblique incidence Energy gap Further Readings: Klevenhagen, S. C. 1985. Physics of Energy level Electron Beam Therapy Medical Physics Handbooks 13, Adam Hilger Ltd, Bristol, U.K; Podgorsak, E. B. 2003. Review of Valence band Radiation Oncology Physics: A Handbook for Teachers and Students, International Atomic Energy Agency, Vienna, Austria; Williams, J. R. and D. I. Thwaites. 2000. Radiotherapy Physics in Hole Practice, 2nd edn., Oxford University Press, Oxford, UK. FIGURE E.31 If an electron gains energy to cross the band gap, it is Electron off axis factor free to conduct in the conduction band. It leaves behind a hole into which (Radiotherapy) The off axis factor (off axis ratio – OAR) repre- valence band electrons can move. sents the ratio of the dose at a point a certain distance from the PDD (%) Electron paramagnetic resonance (EPR) 327 Electron spin beam central axis and the dose at a point on the central axis in electron energy (MeV) at the surface (E–0) using the following the plane perpendicular to the beam axis where both points are equation: placed. Off axis factors can be measured at any depth and a plot of off axis factors against distance from central axis gives the beam æ z ö profile at that depth. Thus, off axis factors follow changes in the Ez = E0 çç1- ÷ è R ÷ beam flatness as a function of the depth and the distance from the p ø central axis. Abbreviation: OAR = Off axis ratio. Related Articles: Electron ranges, Electron maximum range, Related Articles: Beam flatness, Beam symmetry, Off axis Electron therapeutic range dose distribution Electron ranges Electron paramagnetic resonance (EPR) (Radiotherapy) It is possible to characterise electron beams used (General) Electron paramagnetic resonance (EPR) is a spectro- in radiotherapy by the depth at which a certain point on the depth metric method which is based on the resonant absorption of a dose curve is located. Typical values are depth of maximum dose, hyper frequency electromagnetic wave by a paramagnetic spe- 90% dose, 80% dose, 50% dose and the depth at which the dose cies placed in a static magnetic field. The EPR technique makes is reduced to the value of the proportion of contamination x-rays. it possible to measure the concentration of free radicals, in par- There is a series of simple approximations which relate each of ticular those induced by ionising radiation in organic or inorganic these depths to the nominal energy of the electron beam, and materials. these are listed as follows: E Electron positron pair • Depth of maximum dose (mm) is approximately two (Radiation Protection) See Pair production times the beam energy (MeV). • Depth of 90% dose (therapeutic range) (mm) is approxi- Electron practical range mately three times the beam energy (MeV). (Radiotherapy) The electron practical range (Rp) is found by • Depth of 80% dose (cm) is approximately one-third of using a plot of the central axis percentage depth dose curve (see the beam energy (MeV). Figure E.33). First draw a tangent to the depth dose curve at the • Depth of 50% dose (mm) is approximately four times depth of 50% dose. Next extend the Bremsstrahlung contamina- the beam energy (MeV). tion tail back parallel to the depth axis towards the dose axis. • Electron practical range (mm) is approximately five The point at which these two lines meet is the electron practical times the beam energy (MeV). range (depth). The practical range increases with increasing beam energy and can be approximately found by using the following Therefore applying these approximations for the example of a 6 assumption: MeV electron beam will give the following results: Practical range (in mm) is equal to five times the beam energy The depth of maximum dose is at 12 mm, the 90% dose is at (in MeV), i.e. for a 6 MeV beam the practical range should be 18 mm, the 80% dose is at 20 mm, the 50% dose is at 24 mm, and approximately 30 mm. the practical range is at 30 mm. Conversely the practical range (in cm) can also be used to cal- Knowledge of such depths is used when selecting the appro- culate the most probable beam energy at the surface (Ep,0) by the priate energy for a treatment. Other equations can be used, e.g. following relationship: to relate the practical range to the most probable beam energy at the surface or to the mean electron energy at depth (see the article Ep,0 (MeV) = 0.22 +1.98Rp + 0.0025R2 p Electron practical range for detail on these equations). Related Articles: Electron maximum range, Electron practi- It is also possible to calculate the mean electron energy (MeV) cal range, Electron therapeutic range at depth (E–z) from the practical range (cm) and the mean Electron spin (General) Experiments by Otto Stern and Walter Gerlach in 1921 have shown that the electron, in addition to its orbital angular 100 momentum L,⃗ possesses an intrinsic angular momentum. This 90 intrinsic angular momentum is referred to as the spin S ⃗ and is 80 specified by two quantum numbers: s = 1/2 and ms that can take 70 two values (1/2 or −1/2). 60 The electron spin is given as 50 40 30 S = s ( 3 s +1) = and its z component Sz = ms 2 20 Rp 10 0 The spin quantum number is one of four quantum descriptors of 0 10 20 30 40 50 the atom: principal, orbital, magnetic and spin quantum numbers. Depth (mm) The spin quantum number s for an electron is ½, its z component ms can take two values +½ and −½. FIGURE E.33 An illustration of the definition of the electron practical Electron spin explains hyperfine lines in the hydrogen spec- range. trum and the outcome of the Stern–Gerlach experiment. PDD (%) Electron spin resonance 328 Electron volt In the classical conception of spin, the electron is viewed as a Electron therapeutic range ball of charge spinning on its own axis. While this may be a use- (Radiotherapy) The therapeutic range (R90) is defined as the elec- ful visual aid, spin is a quantum phenomenon and experimental tron isodose that covers the distal boundary of the tumour to be observations are not consistent with this simple mechanical view. treated. Typically the electron energy to be used in a treatment is chosen such that the 90% (or 85% depending on local prac- Electron spin resonance tice) dose level beyond the peak is the isodose that reaches this (General) See Electron paramagnetic resonance distal boundary. The therapeutic range increases with increasing energy. A simple approximation for remembering the depth of Electron stopping power 90% dose is that the depth of 90% dose (in mm) is equal to three (Radiotherapy) The stopping power for electrons and positrons times the beam energy (in MeV), i.e. for a 6 MeV beam the depth is different from that of heavy charged particles because an elec- of 90% dose is 18 mm (Figure E.34). tron loses a large fraction of its energy in a single collision with Related Articles: Electron ranges, Electron maximum range, an atomic electron which has equal mass and also because the Electron practical range electron is identical to the atomic electron with which it collides. Quantum mechanics implies that incident and struck electrons cannot be distinguished. The collisional stopping power for elec- Electron transport trons and positrons can be written as follows: (Radiotherapy) Much of the energy transferred to the medium in radiotherapy is through the electrons when an incident pho- ton (causing an ionisation event) releases an electron and the E latter in turn releases secondary electrons. The electrons contrib- -æ dE ö 4pk2e4n é 2 0 mc t t + 2 ù ç ÷ = 2 êln + F± (b)ú è dx b2 øcoll mc ëê 2I ûú ute to energy deposition along the paths taken. At photon beam energies below 6 MeV electron equilibrium is assumed to hold where (AAPM, 2004). This means that the number and energy distri- bution of electrons entering a local volume around the primary 1- b2 t2 ù photon interaction are equal. In addition the dose contribution F- (b) é = ê1+ - (2t +1) ln2 for electrons 2 ë 8 ú of scattered photons dominates over electrons. Above 6 MeV û these assumptions no longer hold and the dose profile deposition is altered due to electron transport. In dosimetry, mathematical and models of energy deposition due to electron transport have been developed (Figure E.35). Further Reading: AAPM Report No. 85. 2004. Tissue inho- F+ ( ) b2 é ù b = ln2 - ê 14 10 4 23 + + + ú 4 2 2 3 for positrons mogeneity corrections for megavoltage photon beams. Medical 2 t + ëê (t + 2) (t + 2) ûú Physics Publishing, Madison, WI. where Electron volt k (General) The electron volt (eV) is a unit of energy. It is the stan- 0 = 8.99 × 109 N m2/C2 z is the atomic number of the heavy particle dard unit to describe x-ray and gamma ray energies in diagnostic e is the magnitude of the electron charge radiology and radiotherapy as well as energies of atomic and sub- n is the number of electrons per unit volume in the medium atomic particles. m is the electron rest mass One electron volt is the energy gained when an electron is swept c is the speed of light in vacuum across a potential difference of 1 V. It is a convenient unit for use β = v/c is the speed of the particle relative to c as the definition of the electron volt can be directly related to the I is the mean excitation energy of the medium basic physics of an x-ray tube or a linear accelerator. For example, τ is the T/mc2 kinetic energy T of the electron or positron if 70 kV is applied across an x-ray tube, electrons are swept across expressed in multiples of the electron rest energy mc2 the tube and strike the anode with a maximum energy of 70 keV. The formula is obtained by combining soft collisions and hard collisions using the Moller’s electron cross section and Bhabba’s positron cross sections for free electrons. 100 The total stopping power for electrons and positrons is the sum 90 of the collisional and radiative stopping powers: 80 70 ± ± ± æ dE ö 60 ç - ÷ = æ dE ç - ö ÷ + æ dE ö dx dx ç - è ø è øcol è dx ÷ ø 50 rad 40 The following approximate formula gives the ratio of radiative 30 R90 and collisional stopping powers for an electron of total energy T, 20 expressed in MeV, in an element of atomic number Z: 10 0

0 10 20 30 40 50 (-dE /dx) rad ZT Depth (mm) (- E dx) » d / 800 col FIGURE E.34 An illustration of the electron therapeutic range. PDD (%) Electronic binding energy 329 Electronic equilibrium Incident Scattered photon photon Scattered Secondary electron electron paths Medium Tape volume FIGURE E.35 Energy deposited by secondary electrons. The maximum x-ray energy in keV is then numerically equivalent E to the maximum voltage across the tube in kV: 1eV = 1.602´10-9 J FIGURE E.36 Electronic equilibrium. Electronic binding energy (General) The binding energy of an electron in an atom is the energy required to remove the electron from the atom. The bind- ing energy is characteristic of the atom and the ‘shell’ from which Range R the electron is removed. Electrons in the orbits closer to the 100 100 100 100 nucleus are more tightly bound and have higher binding energies. For example, the K and L shells of tungsten have binding ener- A B C D E F G gies of 69.5 and 12.1 keV, respectively. In atoms, binding energy Kerma ranges from a few electron volts required to ionise an alkali atom Absorbed dose by removing its outer shell electron to about 150 keV required to Kerma remove a K shell electron from a very high atomic number atom. Build up Electronic region equilibrium Electronic equilibrium (Radiotherapy) The absorbed dose deposition by photons through interaction with matter is a two step process in which they first (a) Depth transfer their energy to electrons (charged particles) which in turn A B C D E F G deposit energy in the matter. However, the energy loss by photons 100 95 90 86 82 at a particular location is not necessarily equal to the absorbed 78 dose by the medium at that location. This is because the second- ary electrons (charged particles) travel a certain distance from Kerma the interaction point before they release all their kinetic energy. Electronic equilibrium exists for a volume v if each electron of Absorbed dose a given energy leaving v is replaced by an electron of the same Kerma Build up In this region there is not energy entering the volume (Figure E.36). region strict electronic equilibrium In the volume where the electronic equilibrium exists Equil. thickness æ m (b) Depth = = en ö D Kc Yç r ÷ è ø FIGURE E.37 Illustrating (a) electronic equilibrium condition and (b) where transient electronic equilibrium. D is the dose Kc is the collision kerma Ψ is the energy fluence the interface, then the dose at the volume considered is due part to μen/ρ is the mass energy absorption coefficient electrons originating in a region of different photon fluence or of different composition. In this case the dose in the volume element If the photon fluence changes over a distance comparable with will not be equal to the energy removed from the radiation beam the range of the secondary electrons in the matter because of atten- in that volume. uation or if the volume considered is near an interface between In Figure E.37a one hundred secondary electrons with a range different media, i.e. within the range of secondary electrons from R are supposed to be set in motion in any square from A to G, Electronic focusing 330 E lectronic portal imaging without any photon beam attenuation. The dose increases with medical. The technology is also referred to as e-ink. The dis- depth to reach a plateau while the photon fluence is constant. played image is viewed by the reflection of the environmental In this situation the electronic equilibrium can be obtained in light and does not emit light itself. This can reduce eye fatigue any point beyond the build-up region. In Figure E.37b a fluence experienced with bright monitor displays. attenuation of 5% will reduce by the same amount the secondary The electronic paper consists of a sheet with transparent micro electrons in any square from A to G. The electronic equilibrium capsules. The capsules vary in size – currently between 10 and cannot be obtained because of the photon attenuation and the sub- 100 µm. Each capsule contains an oily solution with black dye (the sequent variation of Ψ. For depth beyond the build-up region, the electronic ink). Inside the oily solution are numerous white parti- D-curve becomes parallel to the Kc-curve and a condition called cles (titanium dioxide), which are negatively charged. The micro- transient electronic equilibrium is obtained where capsules are held in a layer of liquid polymer, forming a sheet with an approximate thickness of 80 µm. The polymer with capsules D @ K is placed between two arrays of flat electrodes. The upper array is c (1+ m¢x ) transparent (indium tin oxide). where The two arrays of electrodes are aligned in a way to form μ′ is the common slope of the D- and K-curves square pixels. Each pixel is formed by a pair of electrodes situ- x – is the mean distance the secondary electrons carry their ated on both sides of the polymer sheet. The size of the electrodes kinetic energy in the direction of the photons determines the spatial resolution of the display. The network of electrodes is connected to a display circuitry E Electronic focusing that applies DC voltage to specific pairs of electrodes. (Ultrasound) Electronic focusing refers to a combination of Applying a negative charge to the surface electrode repels the ‘dynamic transmit focusing’ and ‘dynamic receive focusing’ – negatively charged white particles to the bottom of the capsules described separately – used in combination with a lens to focus that is between the electrodes. This causes the black dye to appear the beam across the axial (vertical), lateral (width) and elevation on the surface making the pixel dark. They will appear black if (slice thickness) planes. the white particles are pushed to the very bottom of the micro cap- sule. Reversing the voltage attracts the negatively charged white Electronic generators particles to the surface. This gives a pale grey level to the pixel or (Nuclear Medicine) Electronic generators are automated synthe- white if all particles are attracted to the surface. sis systems which are integrated with compact cyclotron technol- Currently the contrast resolution of electronic paper displays ogy for the preparation of radiopharmaceuticals. These are for use (EPD) is very low – not more than 16 grey levels for commercial in preclinical and clinical investigations with positron emission displays. This because it is still very difficult to control the exact tomography. The system is usually controlled by a PC. position of the white particles inside the dark dye in the micro capsule. However, the spatial resolution currently is over 300 dpi. Electronic medical record (EMR) The EPD is very slow, currently requiring to be ‘erased’ with a (Diagnostic Radiology) The electronic medical record (EMR) is constant charge applied to all electrodes as an intermittent stage a digital version of the medical record – a systematic collection of when moving from one image to another. data related to patient’s medical history and care. This type of display is still not suitable for medical imaging, Most common types of data stored under the EMR: but its parameters are constantly improving. EPD consume very small amounts of energy and are suitable for battery powered • Patient demographics equipment – e.g. e-book readers (such as Amazon Kindle). Some • Medical history new electronic paper displays can produce colour images, still • Examination data with limited number of nuances. • Progress reports • Allergy lists Electronic portal imaging • Immunisation status (Radiotherapy) In external beam radiotherapy, portal imaging • Laboratory test results involves using the radiation from the treatment beam (i.e. the • Radiology images (x-rays, CT, MRI, US, etc.) treatment portal) that passes through the patient to form an image. • Other clinical images – endoscopy, laparoscopy, clini- The variations in absorption across the field, particularly between cal photographs bone, soft tissue and the field edge produce the image. • Medication information Film versus Electronic Detectors: Traditionally film has been • Evidence-based recommendations for specific medical used. More recently electronic detectors have been developed and conditions now largely replace film. These have the advantage of ease of use • Administrative data – record of appointments, remind- (e.g. no need to process the image), the information is available ers, billing records immediately and may be analysed automatically. Information Measured: Electronic portal imaging (EPI) EMRs are handled using distributed information systems (IS) is used to monitor the accuracy of treatment delivery, usually within the enterprise (such as hospital information system [HIS], by determining the positioning of bony anatomy relative to the picture archiving and communication system [PACS], etc.). The edge of the radiation field. This may be compared with a pre- application and scope are dependent on the particular IS, applica- dicted image from a treatment simulator or a DRR generated tion, enterprise structure, functional and operational levels, etc. from the TPS. The clinical target volume may move relative to the bony structure and hence the use of bony structures for Electronic paper set-up may result in uncertainty in the positioning of the target (Diagnostic Radiology) Electronic paper (or e-paper) is a type of volume. Figure E.38a shows a DRR for a prostate treatment and flexible display used in various imaging applications including Figure E.38b the corresponding electronic portal image. Electronic portal imaging device 331 Electronic portal imaging device Camera Systems: Some of the first systems to be developed were based on the use of cameras. Figure E.39 shows a sche- matic diagram of such a detector. X-rays transmitted through the patient interact with a fluorescent screen to produce light. The mirror takes the light out of the x-ray beam path to a camera and lens where it is imaged. The whole assembly is in a light-tight box. The mirror is needed to prevent radiation damage to the camera. The camera may be a charge coupled device (CCD) or a tube camera. Liquid Ionisation Chamber Array: One particular design of EPID is based on the use of an array of ionisation chambers located in a box containing iso-octane as the ionised medium. The detector is read out chamber-by-chamber using an array of switches and multiplexers to select the row and column to be read (Figure E.40). Flat Panel Imagers: Most EPIDs used now are based on amorphous silicon flat panel technology. In these detectors a two- dimensional array of radiation sensitive pixels is deposited onto FIGURE E.38A Digitally reconstructed radiograph of prostate a glass substrate. Each pixel has a field effect transistor (FET) treatment. E associated with it that acts as a readout switch (Figure E.41). Figure E.42 shows a photograph of such a device. Transmitted x-rays Fluorescent screen Lens Mirror Camera FIGURE E.38B Corresponding electronic portal image. Light-tight housing Comparison with Diagnostic Imaging: Images for EPI are FIGURE E.39 Schematic diagram of camera-based EPID. generally of lower quality than diagnostic, kV energy images because of the high energy used of several MV Abbreviations: DRR = Digitally reconstructed radiograph, EPI = Electronic portal imaging, kV = Kilovoltage, MV = Megavoltage and TPS = Treatment planning system. Multiplex Related Articles: Electronic portal imaging device, Port film, Amplifier Online portal imaging, Digitally reconstructed radiograph (DRR) Further Reading: Antonuk, L. E. 2002. Electronic portal Electrometers imaging devices: A review and historical perspective of contem- porary technologies and research. Phys. Med. Biol. 47:R31–R65. Electronic portal imaging device Control unit Ionisation (Radiotherapy) An electronic portal imaging device (EPID) is a chamber detector that produces a 2D image of the treatment portal, usually matrix with the patient present in the field. The images obtained are gen- erally of megavoltage energy and thus of lower quality than diag- nostic, kilovoltage energy images. Several technologies have been developed for EPIDs: camera system; liquid ionisation chamber FIGURE E.40 Schematic view of readout of a liquid ionisation chamber arrays and flat panel detectors array. High voltage switches Electroscope 332 Elevation resolution FET object due to the Coulomb electrostatic force. There are various types of electroscopes. The most common type is the gold-leaf electroscope, in which two thin gold leaves are suspended from Data lines a conducting rod held in an insulated container. When a source of static electricity is brought near to the rod, some of the elec- trons

in the rod are pushed to the leaves if the source is negative or pulled up to the rod from the leaves if the source is positive. In both cases the leaves are charged likewise and so they repel each other and separate. If the sources are always held at the same FET lines distance from the rod the amount of separation is proportional to the charge of the source. The charged electroscope can also be used to detect ionising radiation. If a radioactive material is intro- duced into a charged electroscope, the charge on the sensor will be gradually neutralised by the oppositely charged ions formed by the radiation from the surrounding air molecules. The rate of discharge is proportional of the intensity of the radiation present. A type of electroscope is also used in the quartz fibre radiation dosimeter. E Electrostatic deflector (Nuclear Medicine) The electrostatic deflector is used in a posi- Bias lines tive particle cyclotron to redirect the particle beam (typically Sensor protons) onto a target. In a positive particle cyclotron, the elec- trostatic deflector has a negative potential in order to extract the FIGURE E.41 Schematic diagram of structure of an amorphous silicon beam from the magnetic field. flat panel imager. Here a 2 × 2 pixel region is shown. A real device would Related Articles: Cyclotron, External beam irradiation contain 512 × 512 pixels typically. Elementary particles (General) The term elementary particle refers to a subatomic par- ticle with a definite mass and charge which contains no discern- able substructure. Many families and sub-families of elementary particles exist. Three basic quantities are used to identify the particles: mass, charge and spin. According to their masses and the dominant interaction particles are grouped into four families: bosons (graviton and photon, massless bosons), leptons (fermions, light particles), mesons (bosons, intermediate mass), baryons (fer- mions, heavy particles). Fermions, having half-integer spin, obey the exclusion principle and bosons do not because they have inte- ger spin. Baryons and mesons are subject to all four interactions: strong, electromagnetic, weak and gravitational. Leptons are not sensitive to strong interaction while photons (massless bosons) are sensitive to electromagnetic interaction. Elevation resolution (Ultrasound) The elevation plane is the plane perpendicular to the FIGURE E.42 Photograph of an amorphous silicon flat panel imager. scan plane, as shown in the image. The scan plane is more eas- ily understood as that represented by the two-dimensional screen image. The shape and size of the beam in both planes differ with Abbreviations: ADC = Analogue digital converter, CCD = the beam width in the scan plane relating to the lateral resolution Charge coupled device, EPID = Electronic portal imaging device and in the elevation plane to the ‘slice thickness’. and FET = Field effect transistor. The elevation plane exists because the crystal elements within Related Articles: Electronic portal imaging device, Port film, an array are three-dimensional objects, typically rectangular in Online portal imaging, Digitally reconstructed radiograph (DRR) shape with the longer side parallel to the elevation plane. The Further Reading: Antonuk, L. E. 2002. Electronic portal size of the transducer elements is proportional to the wavelength imaging devices: a review and historical perspective of contem- they produce and therefore higher frequency transducers have porary technologies and research. Phys. Med. Biol. 47:R31–R65. smaller dimensions. By extension, the slice thickness also relates to a transducer type, with higher frequency transducers having a Electroscope smaller slice thickness. (General) An electroscope is an early scientific instrument for While the elevation plane is not visible to the operator, it influ- detecting the presence and magnitude of electric charge or ion- ences what the operator sees in particular regarding the spatial ising radiation. The device’s operation is based on the Coulomb resolution. The spatial resolution is the minimum space two sepa- force law and detects the electric charge by the motion of a test rate objects can be determined by the system and is calculated Embryo 333 EMERALD II project using the axial (vertical), lateral (width) and elevation resolution This original project produced one of the world’s first ISBN- (slice thickness). numbered electronic image databases (on CD-ROM). This is also A partial volume artefact can occur when structures are one of the first ISBN-numbered e-learning materials in the world smaller than (or only monopolise part of) the total slice thick- (and the first ISBN-numbered e-Learning material in medical ness. Focusing on the elevation plane therefore reduces the size of physics). The first ISBN-numbered electronic image databases the slice thickness and therefore the potential for partial volume (on CD-ROM, all being in the field of medicine) emerged almost artefacts. simultaneously (within four months) – Figure E.43 The volume of the IDB is about 1400 images of imaging/ radiotherapy equipment and its components; block diagrams and performance parameters, graphs, waveforms; QA procedures and measuring equipment; test objects and image quality examples; typical images and artefacts, etc. Initially the EMERALD e-learning training materials were distributed on CD-ROM and since 2000 they have been accessed mainly via the web site www .emerald2 .eu (see EMERALD II project). Since 2000 the web site has more than 2000 users per month, mainly from low and middle income countries. These e-learning materials were first accessed/made available for exter- nal access in 1996 at the ICTP College of Medical Physics (this Embryo E event marks the first application of e-learning in medical physics). (General) In humans the term embryo is used to describe the Related Articles: EMERALD II project, EMIT Project, developing baby from the moment of fertilisation until the end of e-Learning the 8th week; after this the term used is foetus. Further Reading: Tabakov, S. and V. Tabakova. 2015. The Related Article: Foetus Pioneering of e-Learning in Medical Physics, Valonius Press, London, available free at: www .emerald2 .eu /mep _15 .html EMERALD project (General) European Medical Radiation Learning Development EMERALD II project (EMERALD) is an international project, supported by the EU (General) European Medical Radiation Learning Development Leonardo Programme (1995–1998). The partners in the proj- – Internet Issue (EMERALD II) is an international project, sup- ect were organised as a consortium of universities and hospi- ported by the EU Leonardo Programme (1999–2001). The part- tals from UK, Sweden, Italy and Portugal as follows: King’s ners in the project were organised as a consortium of universities College London, University of Lund, University of Florence, and hospitals from the following: King’s College London; King’s King’s College Hospital, Lund University Hospital, Florence Healthcare Trust; University of Lund; Lund University Hospital; University Hospital, The Portuguese Oncological Institute in Portuguese Institute of Oncology, Lisbon; University of Florence; Lisbon, the High School of Medical Technology Lisbon and Florence University Hospital; Centre Alexis Vautrin – Nancy; the International Centre for Theoretical Physics (ICTP) in Czech Technical University; Northern Ireland Regional Medical Trieste, Medical University Plovdiv (silent partner), with Project Physics Agency; St James’s Hospital, Dublin. Project Manager Contractor: V. C. Roberts, Coordinator: S. Tabakov and repre- was Prof. C. Roberts, and project Coordinator Dr S. Tabakov. sentatives of the partners: S.-E. Strand, B.-A. Jonsson, J. Gomes Representatives of the projects partners were all those included in da Silva, F. Milano, C. A. Lewis, I.L. Lamm, A. Benini and A. the EMERALD project and additionally: A. Noel, L. Musilek, P. Litchev. Smith, N. Sheahan and S. Bowring. EMERALD project developed the first structured training The EMERALD II project developed the first educational web with e-learning in medical physics. It is organised in three mod- site in medical physics (1999) – www .emerald2 .net, later this web- ules – Physics of: X-ray Diagnostic Radiology, Nuclear Medicine, site was also available at www .emerald2 .eu. The web site hosts Radiotherapy. Each module of 80 days includes a workbook with the e-learning materials of projects EMERALD and EMIT, plus training tasks and image database (IDB). a number of materials related to medical physics education and FIGURE E.43 The EMERALD CD – one of the three first ISBN-numbered e-learning materials. Emission computed tomography (ECT) 334 End diastolic velocity training. Since 2000 the web site has a steady number of visitors In 2004 the EMIT project was awarded the EU inaugural – more than 2000 users per month, mainly from low and middle award for vocational education – Leonardo da Vinci Award. income countries. The initial design of the web site was created Related Articles: EMERALD project, EMERALD II Project, by M. Ljugberg and B.-A. Jonsson, later the design was handled EMITEL project, e-Learning by M. Stoeva and A. Cvetkov. Further Readings: Tabakov, S., P. Sprawls, A. Krisanachinda The project disseminated the use of e-learning in medical and C. Lewis. 2011. Medical Physics and Engineering Education physics and organised 14 seminars on the subject (in Europe and and Training – Part I, ICTP, Trieste, Italy, ISBN 92-95003-44-6; USA). Tabakov, S. and V. Tabakova. 2015. The Pioneering of e-Learn- Related Articles: EMERALD project, EMIT Project, ing in Medical Physics, Valonius Press, London, available free at: e-Learning www .emerald2 .eu /mep _15 .html Further Reading: Tabakov, S. and V. Tabakova. 2015. The Pioneering of e-Learning in Medical Physics, Valonius Press, EMITEL project London, available free at: www .emerald2 .eu /mep _15 .html (General) European Medical Imaging Technology e-Encyclopae- dia for Lifelong Learning (EMITEL) is an international project, Emission computed tomography (ECT) initially supported by the EU Leonardo Programme (2006–2009). (Nuclear Medicine) Tomography is a method to reconstruct The partners in the project were organised in a consortium of uni- transversal images from projections of data acquired in different versities and hospitals as follows: King’s College London; King’s angles around the patients. The reconstruction is usually made by Healthcare Trust; University of Lund; Lund University Hospital; E filtered backprojection or iterative methods. In nuclear medicine University of Florence; AM Studio, Plovdiv, and the International measurements are made with either radionuclides that emit single Organization for Medical Physics (IOMP). This was the first photons or with positron emitters that produce two 511 keV pho- EC project for IOMP. The project manager and coordinator for tons from positron annihilation. These two imaging modalities EMITEL was S. Tabakov. The project is the largest in the pro- are referred to as single-photon emission computed tomography fession with over 300 experts from 36 countries taking part in (SPECT) and positron emission tomography (PET). The fam- its creation. The initial EMITEL team continued to function five ily name for these modalities is emission computed tomography years after the end of the project as a network (coordinated by V. (ECT). Tabakova). The EMITEL project developed the first e-Encyclopaedia EMIT project of Medical Physics and a multilingual e-Dictionary of Medical (General) European Medical Imaging Technology Training Physics Terms (translated to 30 languages). Both reference mate- (EMIT) is an international project, supported by the EU Leonardo rials are hosted at a special web site (designed by M. Stoeva, A. Programme (2001–2004). The partners in the project were organ- Cvetkov and S. Tabakov). The web site www .emitel2 .eu (also ised in a consortium of universities and hospitals as follows: www .emitel2 .net) was launched in 2009 and since this time has King’s College London; King’s College Hospital Healthcare a steady number of over 5000 visits per month from all over the Trust; University of Lund; Lund University Hospital; University world. The materials are of special importance for medical phys- of Florence; Florence University Hospital; Hospital Albert ics education. Michallon, Grenoble; the European Federation of Organisations This e-encyclopaedia was further developed and updated by for Medical Physics (EFOMP), International Centre for Theoretical a team of specialists who volunteered their contribution for the Physics (ICTP), Trieste, and Tempus Medical Radiation Physics global development of the profession. The result formed the basis Centre ERM in Plovdiv (part of Medical University Plovdiv). This of the current free online Encyclopaedia of Medical Physics. The was the first EC project for EFOMP. The project manager was initial version of the encyclopaedia was published by CRC press C. Roberts, and project coordinator S. Tabakov with representa- in 2013, while the current updated 2nd Edition is published by tives of the project partners – also all listed in the EMERALD CRC Press in 2021. and EMERALD projects as A. Simmons, S. Keevil, V. Tabakova, Related Articles: EMERALD project, EMERALD II Project, C. Dean, D. Goss, V. Aitken, R. Wirestam, F. Stahlberg, M. EMIT project, e-Learning Almqvist, T. Jansson and

J.-Y. Giraud. Further Readings: Tabakov, S., S.-E. Strand, F. Milano, C. Using the experience of developing the EMERALD and Lewis and P. Sprawls. 2013. Editors in Chief (with 108 co-authors) EMERALD II projects, and their website space, the EMIT project of the EMITEL, Encyclopaedia of Medical Physics, CRC Press developed two training modules (on MRI and ultrasound medi- (Taylor & Francis Group), ISBN 978-1-4665-5550-1 (vol. I); ISBN cal imaging), each lasting 80 days. Each module is supported by 978-1-4665-5555-6 (vol. II), New York, NY, and Boca Raton, FL; a workbook with training tasks and image database (IDB). The Tabakov, S., V. Tabakova, M. Stoeva, A. Cvetkov, F. Milano, S.-E. volume of the IDB is about 1700 images. The modules are sup- Strand, J.-Y. Giraud and C. Lewis. 2013. Medical physics thesau- ported by a guide for the users and trainers, including a number of rus and international dictionary. J. Med. Phys. Int. 2:139–144; sample documents for clinical training organisations. All EMIT Tabakov, S. and Tabakova, V. 2015. The Pioneering of e-Learning e-learning materials are available at www .emerald2 .eu (also www in Medical Physics, Valonius Press, London. UK, available free .emerald2 .net). In 2003 the EMIT project organised the first inter- at: www .emerald2 .eu /mep _15 .html. national conference on e-learning in medical physics. The EMIT project also developed the foundation of the Emulsion layer Multilingual Dictionary of Medical Physics Terms. It was ini- (Diagnostic Radiology) See Film emulsion tially hosted on CD-ROM and distributed globally at World Congress in Sydney 2003. Later the dictionary was associated End diastolic velocity with the EMITEL project (www .emitel2 .eu, www .emitel2 .net) (Ultrasound) Diastole describes the relaxation of the chambers of and expanded to over 30 languages. Since 2004 the EMIT materi- the heart. The end diastolic velocity is the velocity in the systemic als have over 2000 users per month from all over the world. arterial circulation at the end of ventricular diastole immediately End of bombardment (EOB) 335 E nergy gap The energy absorption coefficient, as defined earlier, is often called linear energy absorption coefficient and has dimension L−1 (usually given in cm−1). It depends on the radiation energy, on the composition and on the density of the medium. PSV The product (μa * I) is equivalent to the energy absorption rate and can be used to calculate the absorbed dose in the medium. Notwithstanding, the absorbed dose is more directly related to the mass energy absorption coefficient which is more often employed EDV in that kind of calculation. Related Articles: Absorbed dose, Absorption coefficients, Mass energy absorption coefficient Energy deposition (Radiation Protection) At a cellular level, damage is caused to the human body by the deposition and absorption of the energy inher- PSV ent in an incident radiation beam, whether it be the energy of ion- ising or non-ionising photons, or the kinetic energy of the atomic particles (alpha, beta, protons, neutrons, etc.) in particle radiation. The nature and severity of damage caused to individual cells, E and to the tissue or organ made up of those cells, can be bet- EDV ter understood by modelling the concentration and distribution of energy deposition within the cells. Such modelling helps to definition any radiation as having either a high linear or low lin- ear transfer characteristic, with associated radiation weighting factor. As the deposition of energy in matter has statistical fluctua- tions, energy deposition is stochastic. Thus energy deposition within a mass of material may be estimated by averaging observed FIGURE E.44 The end diastolic velocity (EDV) in a monophasic and values of the stochastic variations. triphasic arterial waveform. The energy deposited in a single interaction, i, can be defined at the point of interaction: preceding the onset of systole. In the case of a flow in an artery ei = ein - eout + Q leading to high terminal resistance, this value may be zero (Figure E.44). where Abbreviations: EDV = End diastolic velocity and PSV = Peak ɛin is the energy of the incident ionising particle (excluding rest systolic velocity. energy) (expressed in Joules or electron-volts (eV)) Related Article: Peak systolic velocity ɛout is the sum of the energies of all particles leaving the inter- action (excluding rest energy) Q is the change in the rest energies of the nucleus and of all End of bombardment (EOB) particles involved in the interaction (Nuclear Medicine) End of bombardment (EOB) refers to the moment when irradiation stops, i.e. when the beam is turned off on particle accelerators such as cyclotrons. Q can be either +ve or −ve, indicating either a decrease or an Related Article: Avogadro’s number increase, respectively, in the net rest energy; of all the particles Further Reading: Helus, F. 1983. Radionuclides Production, resultant from the interaction Vol. 1, CRC Press, Boca Raton, FL. Related Articles: Absorbed dose, Linear energy transfer, Radiation weighting factor Energy absorption coefficient Energy fluence (Radiation Protection) The energy absorption coefficient, μa, (Radiation Protection) Energy fluence, usually denoted by the is the fraction of the energy of a gamma or x-ray beam that is Greek letter ψ, is one of the units used to describe a radiation absorbed per unit distance in the medium. Since a fraction of the field. It is the radiance (product of number of particles, N, and incident beam is scattered, the absorbed energy is only a part of particle energy, E) per unit area. the total energy lost by the beam: Hence I = I0 ×exp éë-(ma + ms ) tùû dR Y = da where I0 is the incident radiation intensity at the surface of the The unit of energy fluence is J/m2. medium Related Article: Incident energy fluence I is the radiation intensity transmitted through a medium of thickness t Energy gap μs is the fraction of the beam scattered per unit distance (Nuclear Medicine) See Band gap Energy loss rate 336 ENETRAP project Energy loss rate collimates most of the scattered beam from the degrader), (Radiation Protection, General) The rate of loss of energy from nickel energy slits and collimator (which reduce the energy a beam of ionising radiation as it traverses an absorbing medium spread), and a nickel beam stop (which is used to tune the is described by the linear energy transfer of the type of radiation beam) (PTCOG, 2010). Energy degraders are problematic from and the absorption characteristics of the medium. a radiation protection point of view because they produce neu- Related Articles: Absorption, Attenuation, Linear energy trons, which require appropriate shielding. Induced radioactiv- transfer ity of the metal slits/collimators is also significant (Khan and Gibbons, 2014). Energy resolution The reader is referred to IPEM Report 75 for a schematic dia- (Radiation Protection) Energy resolution is the ability of the gram of an ESS. radiation detector of a measuring system (e.g. spectrometer) to Related Articles: Double scattering, Cyclotron, Synchrotron discriminate photons and particles with different energies. Further Readings: Horton, P. and D. Eaton. 2017. Design and The energy resolution (in %) is defined by the full width at half Shielding of Radiotherapy Treatment Facilities, IPEM Report maximum (FWHM) of a pulse (photopeak) divided by the energy 75, 2nd edn., IOP Publishing; Khan, F. M. and J. P. Gibbons. at the maximum of the pulse (Figure E.45): 2014. Khan’s the Physics of Radiation Therapy, 5th edn., Wolters Kluwer Health; Mukherjee, B. 2012. Radiation safety issues relevant to proton therapy and radioisotope production Energy resolution (%) = 100´ FWHM/Emax medical cyclotrons. Radiat. Prot. Environ. 35:126–134; PTCOG E Particle Therapy Co-Operative Group. 2010. PTCOG Report Semiconductor detectors (e.g. Germanium detector) have very 1, Shielding Design and Radiation Safety of Charged Particle narrow pulses, and their energy resolution is very good (below Therapy Facilities, PTCOG Publications Subcommittee Task 1%). This makes them very suitable for spectrum analysers. Group. Scintillation detectors have energy resolution of the order of 8%–9%. Abbreviation: FWHM = Full width at half maximum. Energy spectrum Related Article: Multichannel analyser (Diagnostic Radiology) The energy spectrum is the range and dis- tribution of photon energies in an x-ray beam. Energy selection system See also Beam spectrum. (Radiotherapy) High-energy proton beams can be generated by Related Article: Beam spectrum, Spectrum, discrete either a cyclotron or a synchrotron. Clinical cyclotrons typically generate proton beams of up to 250 MeV (Mukherjee, 2012). ENETRAP project (European Network on Education and A cyclotron uses a strong magnetic field to confine a beam of Training in Radiological Protection project I, II, III) accelerated protons within a circular orbit. Protons are injected (General) The overarching objective of these three consecutive at the centre of a cyclotron and accelerated in a circular path by EU-funded ENETRAP projects was principally to enhance the a high frequency accelerating voltage (applied across two hol- education and training of radiation protection experts (RPE) in low metal D-plate electrodes, placed in the strong perpendicular Europe. RPE are concerned with the protection of workers and magnetic field). On each orbit the protons receive increments of the general public (as opposed to medical physics experts who are energy which increases the radius of their orbit until they reach involved principally with the protection of patients, carers and the periphery of the D-plates. They are then extracted (PTCOG, comforters and volunteers in medical research). In practice given 2010). their extensive expertise in radiation physics MPE often acquire The proton beam extracted from the cyclotron is passed to the necessary knowledge, skills and competences to act also as an energy selection system (ESS). An ESS consists of a graph- RPE in the medical area. ite wedge-shaped energy degrader (which scatters the protons The specific objectives of the ENETRAP projects were: and increases the energy spread), a tantalum collimator (which a. To better integrate existing education and training activities in the radiation protection infrastructure of European countries in order to combat the decline in both student numbers and teaching institutions. b. To develop more harmonised approaches for education 1 and training in radiation protection in Europe and their implementation. c. To better integrate the national resources and capacities for education and training. d. to provide the necessary competence and expertise for 0.5 the continued safe use of radiation in industry, medicine FWHM and research. These objectives were achieved by the establishment of a European education and training network in radiation protection which: Energy a. Assessed training needs and capabilities. b. Identified the potential users and their future involve- FIGURE E.45 Measuring of energy resolution of an energy spectrum ment in order to insure the sustainability of the (photopeak). network. Relative intensity Enhancing filter 337 Ensemble length c. Launched a consortium of universities with the aim of The partners for ENETRAP III were: creating a European Master’s in Radiation Protection. • European Federation for Organisations for Medical d. Reviewed the scientific contents of existing education Physics (EFOMP) and training activities. • Stichting EUTERP, Netherlands e. Explored the effectiveness of on-the-job training and • Department of Health, United Kingdom identified options for additional programmes. • Federal Office for Radiation Protection (Bundesamt für f. Proposed recommendations for the recognition of Strahlenschutz), Germany courses and competences of radiation protection • Commissariat a l‘Energie Atomique et Aux Energies experts. Alternatives, France g. Made recommendations for revising the European • Karlsruher Institut für Technologie,Germany Radiation Protection Course to include a system for • Centro de Investigaciones Energeticas, credit points and modern educational tools, such as dis- Medioambientales y Technologicas, Spain tance learning. • Nuclear Research and Consultancy Group, Netherlands • Associacao do Instituto Superior Tecnico Para a The main deliverables of the ENETRAP project were: Investigacao e Desenvolvimento, Portugal a. The establishment of a universities’ consortium. • Budapesti Muszaki es Gazdasagtudomanyi Egyetem, b. The delivery of pilot modules for training in radiation Hungary protection (including specialised modules for RPE, • PGE Polska Grupa Energetyczna SA, Poland such as the medical and waste management areas). • Université de Lorraine, France E c. The recommendations to the EUTERP platform (http:// www .euterp .eu) regarding the recognition of this train- Hyperlinks: ENETRAP project: www .euterp .eu /staticPage ing, especially for radiation protection experts. .asp ?pageID =40; ENETRAP I: www .e uterp .eu /s tatic Page. asp ?p ageID =40 &s ubpag eID =1 54; ENETRAP II: www .e uterp .eu /s tatic Page. asp ?p ageID =40 &s ubpag eID =1 62; ENETRAP III: The projects were coordinated by the:

www .e uterp .eu /s tatic Page. asp ?p ageID =40 &s ubpag eID =1 63; • Belgian Nuclear Research Centre, Belgium (SCK-CEN) ENETRAP page at SCK-CEN: https://enetrap .sckcen .be /en Dr. Michèle Coeck (mcoeck @sckcen .be) Enhancing filter The main project partners for ENETRAP I were: (General) An enhancing filter is any device which alters the data • The Institute for Nuclear Sciences and Technology, being passed through in such a way as to enhance or make more France prominent the information required, whilst reducing or limiting • Forschungszentrum Karlsruhe, Germany other data. • Federal Office for Radiation Protection (Bundesamt für Enhancing filters may be made of a physical layer of material Strahlenschutz), Germany through which the signal passes (e.g. optical or x-ray) on its way • The Italian National Agency for New Technology, to the detector. Optical filters may enhance specific colours or Energy and Environment, Italy limit resolution, whilst x-ray enhancing filters may be made out • The Nuclear Research and Consultancy Group, The of appropriate metals to harden or soften x-ray beams, to better Netherlands image specific tissues. • The Research Centre for Energy, Environment and Enhancing filters may also be used to post-process digital Technology, Spain images. Many common image enhancement software pack- • The Health Protection Agency, United Kingdom ages exist for enhancing photographic images, though in medi- • Université Joseph Fourier, France cal image applications, more specific software is usually used. • North Highland College, Scotland Typical enhancement processes include edge enhancement, noise removal and equalisation filters. The partners for ENETRAP II were: • Commissariat à l'Energie Atomique, Institut National Ensemble length des Sciences et Techniques Nucléaires, France (Ultrasound) The term ensemble length is used to describe the • Forschungszentrum Karlsruhe, Centre for Advanced number of pulses used to produce each colour ‘line’ in a colour Technological and Environmental Training, Germany flow image. Colour flow systems typically use autocorrelation • Federal Office for Radiation Protection (Bundesamt für methods to compare the shift in phase between pulses and thereby Strahlenschutz), Germany the Doppler frequency. SNR and therefore the quality and accu- • Italian National Agency for New Technology, Energy racy of the colour flow output generally improve with increased and Environment, Italy ensemble length. However, increasing ensemble length leads to • Nuclear Research and Consultancy Group, The time constraints with a possible reduction in frame rate. Netherlands In some systems, ensemble length may be altered by the • Spanish Research Centre for Energy, Environment and operator to improve the sensitivity of the colour flow image at the Technology, Spain expense of frame rate. The converse is also possible. The nomen- • Health Protection Agency, United Kingdom clature of this control varies from scanner to scanner. Ensemble • European Nuclear Education Network Association, France length is one of the many parameters, including line density, • Nuclear and Technological Institute, Portugal frame rate colour box width and depth and pulse repetition fre- • Budapest University of Technology and Economics, quency which are varied automatically to maintain acceptable Institute of Nuclear Techniques, Hungary frame rate if changes are made to one or more of the others by • University Politehnica of Bucharest, Romania an operator. Entrance dose 338 Equalisation Further Reading: Cobbold, R. S. C. 2007. Pulsed methods for flow estimation and imaging, Chapter 10. In: Foundations of Biomedical Ultrasound, Oxford University Press, Oxford, UK. Entrance dose (Radiotherapy) Entrance dose is the quantity which is often mea- sured in the process of in vivo dosimetry. Within in vivo dosim- etry, doses derived from the signal of the detector placed on the skin are compared with the theoretical values calculated by the treatment planning system (TPS). As the accuracy of the skin dose calculation is questionable, and in many cases irrelevant, the signal of the detector is related to the dose at a point which is still close to the skin, but at a certain depth where the accuracy of the TPS is much more satisfactory. At the entrance side of a medium irradiated by a single beam, the dose gradually increases from a low value at the surface up to a maximum value D plified schematic of a conventional bunker design entrance at a depth FIGURE E.46 Sim dmax which depends upon the energy, the collimator opening, the with a simple single stage maze corridor. skin-source-distance, the introduction of beam modifying devices E and the distance separating them from the patient skin, etc. The measurement of Dentrance must be carried out with enough material A typical experimental set-up might measure ESAK with in front and around the detector placed at skin level in order to an ionisation chamber at the position where the surface of the establish the electron equilibrium. body of the patient would be in an x-ray exposure, from which Abbreviation: TPS = Treatment planning system. and entrance surface dose (Gy) and a skin dose (Sv) could be cal- Related Articles: Exit dose, Build-up, Build-up region, Diode culated using Monte Carlo–based conversion factors to take into detectors, In vivo dosimetry, Charged particle disequilibrium account the backscatter from the underlying tissues of the body. Further Reading: Vam Dam, J. and G. Marinello. 2006. Related Articles: Air kerma, Entrance surface dose, Skin dose Methods for in vivo dosimetry in external radiotherapy, ESTRO Booklet No. 1, ESTRO, Brussels, Belgium. Entrance window (Diagnostic Radiology) See Image intensifier Entrance maze EPI (echo planar imaging) (Radiotherapy) Deciding on appropriate bunker shielding for a (Magnetic Resonance) See Echo planar imaging (EPI) radiotherapy accelerator is a complex process. It is vital to know the characteristics of the radiation produced by the accelerator Epidermis with regards to its energy and fluence. (Non-Ionising Radiation) The epidermis is the most superficial Access to the bunker can be by means of a directly shielded layer of the skin, made of an external layer (stratum corneum) door or alternatively a door with little / no shielding combined made of dead keratin-filled cells and an inner layer of live cells with an entrance maze. If a directly shielded door is used it must such as pigment cells (melanocytes) and immune cells. The par- have the same attenuation properties as the adjacent walls, which ticular composition of the dermis varies according to the specific means that the door must be mechanically operated because of its anatomical site, and function. heavy weight. Related Articles: AORD, Basal cell carcinoma, Melanoma, An entrance maze consists of a shielded corridor with a num- UV light hazard, UV dosimetry ber of bends. Particle scatters and absorptions along/within the Further Readings: Blumenberg, M. 2018. Human Skin maze walls reduce the level or radiation that escapes. Cancers: Pathways, Mechanisms, Targets and Treatments, Nonethless, residual dose at the entrance to a maze can stem London, UK; Montagna, W. and P. F. Parakkal. 2012. The from: Structure and Function of Skin, 3rd edn., Elsevier, New York, NY and London, UK. 1. Primary scatter from the bunker and maze walls. 2. Patient scatter EPISTAR 3. Leakage along the maze See Echo-planar imaging and signal targeting with alternating 4. Leakage through the maze walls radiofrequency The dose rate at the maze entrance often includes a contribu- Equalisation tion from neutron radiation. The thickness and composition of (Diagnostic Radiology) Equalisation is a general term for modi- maze walls and door are calculated so that the exposure in adja- fying signals or images to provide a corrected or more equal rep- cent areas outside the bunker are kept below the required safe resentation of the original data. threshold expressed as the dose equivalent per week (Sv/week) In time-varying signals such as sound recordings, equalisation (Figure E.46). refers to the adjustment (in frequency and phase) of the signals to correct for the distortion in their recording, transmission or Entrance surface air kerma (ESAK) storage. (Radiation Protection) The air kerma (Gy) measured at the sur- In images, equalisation may be used to correct for imperfect face of the body. ESAK is useful in relation to estimating entrance ‘illumination’ or to optimise illumination and contrast such as surface doses, and hence skin doses, to patients. through the use of ‘histogram equalisation’. Equilibrium absorbed dose constant 339 Equivalent blur In certain x-ray imagers (e.g. mammography) ‘Scanning blur as illustrated. When blurs from different sources within an Equalisation’ may be used, which delivers a non-uniform distri- imaging procedure are represented with equivalent blur values bution of x-ray exposure so as to maintain optimal sensor bright- they can be directly compared and adjusted to optimise image ness and contrast over the area being imaged. quality. Equivalent blur is defined as a blur with uniform distribution Equilibrium absorbed dose constant that has the same general effect on image quality as a specific blur (Nuclear Medicine) This parameter states the amount of radia- that is being considered. Depending on the shape and intensity tion energy emitted from the radionuclide in the source organ. distribution of a blur, the equivalent blur can be either larger or The energy emitted per unit of cumulated activity is called the smaller than the actual blur dimension. The concept of equiva- equilibrium absorbed dose constant ‘Δ’. This factor needs to be lent blur and methods for determining values were introduced by calculated for each radiation type and it is given by Sprawls in 1977 (Ref.1) (Figure E.47). In radiography there are three major sources of equipment Di = 1.6´10-13 NiEiGy × kg / Bq × s (J) (E.3) related blurring, the x-ray tube focal spot, the absorbing layers within the image receptor especially intensifying screens and where digitising an image into individual pixels. Each pixel is a blur. Ei is the average energy (MeV) of the ith emission There are specific methods for determining the value of the equiv- Ni is the number of particles or photons emitted per alent blur for each source. disintegration For digitised images the dimension of each pixel is a good rep- resentation of the equivalent blur value and be calculated directly E from the field of view (FOV) and matrix size. This gives the pixel The product between the cumulated activity and the equilib- size in the image. geometry and the magnification can be used to rium absorbed dose constant is the total energy radiated by the ith calculate the effective pixel size at the location within the patient emission while activity is present in the source organ. body where it can be compared to other blur values. Another step to determine the absorbed dose to a target organ When focal spot size is measured with the conventional star is to calculate the absorbed fraction (see separate article). The test pattern the resulting value is an equivalent blur size that is absorbed fraction and the equilibrium absorbed dose constant generally different from the actual physical size. For most focal are often combined into a mean dose per cumulated activity S to spots the equivalent size is larger than the actual physical size simplify the calculation procedure. S is determined by the emis- because of the effect of the shape and distribution of the focal sion type, radiation energy and anatomic relationship and is deter- spot. Using the equivalent focal spot size and geometry the equiv- mined for each source-target pair and radionuclide. alent blur can be calculated at the object location where it can be Related Articles: MIRD formalism, Absorbed fraction, directly compared to the blur values from other sources. Cumulated activity, Mean dose per cumulated activity There is a specific method for determining the equivalent blur Further Readings: Cherry, S. R., J. A. Sorenson and M. E. from receptor components, especially intensifying screens and Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, film combinations. It is based on the theory of ‘equivalent pass- Philadelphia, PA, pp. 411–412; Loevinger, R. F., T. Budinger and E. band’ that can be used to calculate the dimension of a uniformly E. Watson. 1988. MIRD Primer for Absorbed Dose Calculations, distributed equivalent blur that would have the same general The Society of Nuclear Medicine Inc., New York; Weber, D. A., effect on image quality as an actual blur. It relies on some shared K. F. Eckerman, L. T. Dillman and J. F. Ryman. 1989. MIRD: characteristics of MTF curves for a range of receptors and the Radionuclides Data and Decay Schemes, The Society of Nuclear relationship to the MTF for a

uniform blur. For a specific receptor Medicine, New York. the equivalent blur (B) can be calculated from the MTF by: Equilibrium dose distribution B = 0.61 / f50 (Radiation Protection) This can be best understood in reference to radiotherapy. The dose that is prescribed to the target area will be equal over that particular area and then decrease around the area, it is important that the distribution of the dose over the tar- get area is equal to achieve optimal results, and this is known as equilibrium dose distribution. This can be achieved by using the multi-leaf/step collimators and wedges. Equivalent blur (Diagnostic Radiology) There are various sources of blurring in all medical imaging procedures. The effect on visibility and image quality depends on the dimension of a blur in relation- ship to the size of objects being imaged. Blurring from differ- ent sources within an imaging procedure such as radiography has different shapes and intensity distributions as illustrated. Because of this the effects on image quality cannot be quanti- tated and compared using the actual physical dimensions. The effective size of a blur with respect to image quality can be dif- ferent from the actual physical size depending on the shape and distribution of the blur. This characteristic can be represented by a profile (point-spread or line-spread function) or an image of the FIGURE E.47 Indicative diagram of blur in radiography. Equivalent dose (HT) 340 Equivalent square Where f50 is the spatial frequency relating to 50% contrast on the Equivalent field size MTF. (Radiotherapy) Equivalent field size is a concept used to deter- Example equivalent blur values published by Kodak for two mine the size of the square radiation beam that has the same film-screen radiographic receptors are 0.44 mm for general radi- radiation dose output and percentage depth dose characteristic as ography and 0.14 mm for mammography. This illustrates the gen- a non-square radiation beam. eral range of blur values in radiography. For further information see Equivalent square. A major value of using equivalent blur is that the blur from Related Article: Equivalent square different sources (focal spot Bfs, receptor Br and pixel Bp) can be directly compared and used to determine the total blur (Bt) for Equivalent mass of radium an imaging process and potentially optimise the imaging factors (Radiotherapy, Brachytherapy) Calibration of source strength is using the following relationship: a very important part of a comprehensive brachytherapy quality system. The instruments, ion-chambers and electrometers, used B for source strength determinations, should have calibrations that t = B 2 B 2 B 2 fs + r + p are traceable to national and international standards. While equivalent blur is especially useful for evaluating image Specification of Source Strength for Photon Emitting quality in radiography the concept can be applied to other imag- Sources: Source strength for a photon emitting source can be ing modalities. It is especially useful for comparing the effect of given as a quantity describing the radioactivity contained in the pixel/voxel size to other sources of blurring within an imaging source or as a quantity describing the output of the source: E process. Further Reading: Sprawls, P. 1977. Equivalent Blur of 1. Specification of contained activity Intensifying Screens. SPIE Vol. 127 Optical Instrumentation in a. Mass of radium; mg Ra Medicine VI. b. Contained activity; Ci, Bq 2. Specification of output Equivalent dose (HT) a. Equivalent mass of radium; mg Ra eq (Radiation Protection) Equivalent dose is a unit used in radiation b. Apparent activity protection. It is the product of the absorbed dose D (in Grays) to c. Reference exposure rate an organ or tissue and a factor, WR called the radiation weighting d. Reference air kerma rate factor, a dimensionless quantity that characterises that damage e. Air kerma strength associated with the relative biological effectiveness of different types of radiation. When brachytherapy was introduced as a treatment modality, the only sources available were radium sources. Source strength was HT = D ´WR given as the mass of the radium contained in the encapsulated source. The filtration of the encapsulation was also given; usu- Equivalent dose has the unit of the Sievert (Sv). ally 0.5 mm Pt for needles and 1–2 mm Pt for tubes. (The United The radiation weighting factor reflects the relative probability Kingdom’s National Physical Laboratory (NPL), a standards/ of damage to DNA arising from the same absorbed dose from measurement institute established at the turn of the last century, different types of radiation due to the density of ionisation events acquired its first radium standard in 1913, made by Marie Curie, along the track of the initiating photon or particle and subsequent and specified in terms of mass of radium.) energy transfer to the tissue. The following table gives the current When artificial radionuclides became available, the brachy- weighting factors: therapy community aimed at specifying source strength for these ‘radium substitutes’ in a radium-like manner. The new sources were similar in shape and strength to the old ones, and thus all the experience gained from the earlier radium treatments could Type of Radiation wR easily be transferred. Photons 1 The equivalent mass of radium, the mgRaEq, for an encapsu- Electrons, betas and muons 1 lated photon emitting source is the mass of Ra-226 filtered by 0.5 Protons 2 mm Pt that gives the same air kerma rate or exposure rate at the Alpha particles and heavy nuclei 20 same distance from the centre of the source. (Note that this could Incident neutron Continuous function – See ICRP lead to interpretation problems for Ra sources. Consider a Ra tube Report 92 with strength 20 mg and filtered by 1 mm Pt, not 0.5 mm Pt; the strength of the tube will correspond to 18.7 mgRaEq.) In modern brachytherapy dosimetry, reference air kerma rate Therefore 1 Gy absorbed dose to a tissue from x-rays or beta par- or air kerma strength is the quantity used to calculate absorbed ticles gives rise to 1 Sv equivalent dose to the tissue. dose. Regulatory dose limits set for individual organs or tissues are See Source strength for a full description of specification of given in terms of equivalent doses and are designed to ensure that source strength. the equivalent doses received by the organs or tissues are below Related Articles: Source strength, Mass of radium, Contained the thresholds for deterministic effects. activity, Apparent activity, Reference air kerma rate (RAKR), Air Related Articles: Absorbed dose, Radiation weighting factor, kerma strength Deterministic effects, Relative biological effectiveness Further Reading: Relative Biological Effectiveness, Radiation Equivalent square Weighting and Quality Factor. ICRP Publication 92, Ann. ICRP (Radiotherapy) Depth dose data is generally only given for square 33(4), 2003. field sizes, as it is impractical to measure every single possible Equivalent tissue air ratio (ETAR) 341 Erect field size and shape. To obtain values for rectangular or irregu- absorbed dose in a patient irradiated by beams of x or gamma rays lar fields, the field shape in question needs to be correlated with in radiotherapy procedures, ICRU Report 24, ICRU, Washington, an equivalent square (or an equivalent field), a shape that has the DC, pp. 22–23; Sontag, M. R. and J. R. Cunningham. 1978. The same scatter/primary ratio (SAR) at the point in question. equivalent tissue-air ratio method for making absorbed dose cal- During the commissioning of a linear accelerator, depth dose culations in a heterogeneous medium. Radiology 129(3):787–794. data for rectangular fields should be measured twice using dif- ferent jaw pairs to define the field. This is because the machine Equivalent uniform dose output may be influenced by the upper or lower jaw position – the (Radiotherapy) Equivalent uniform dose (EUD) is a concept used Collimator exchange effect. to represent a non-uniform dose distribution in terms of a homo- Circular Fields: Circular fields have the same SAR as a geneous distribution. square field of the same area, which roughly equates to an equiva- The following excerpt is from AAPM report 166: lent square of 90% of the diameter of the circular field. The concept of equivalent uniform dose (EUD) proposed by Rectangular Fields: Rectangular fields generally give smaller Niemierko (Niemierko 1997) provides a single metric for report- depth doses than a square field of the same area due to the greater ing non-uniform tumor dose distributions. It is defined as the uni- attenuation of the scatter from the extended sides of the rect- form dose that, if delivered over the same number of fractions angle. Hence to obtain the equivalent square size, the ratio of as the non-uniform dose distribution of interest, yields the same area to perimeter is normally equated. This formulism is known radiobiological effect. To extend the concept of EUD to normal as Sterling’s approximation, and is described by the formula in tissues, Niemierko (1999) proposed a phenomenological formula Equation E.4. This approximation becomes less reliable for large referred to as the generalized EUD, or gEUD: E fields (a,b > 20 cm), and more accurate values can be found in the BJR supplement: 1 æ ö a gEUD = ç a ÷ çåviDi ÷ (E.5) 2ab c = ( (E.4) è i ø a + b) where vi is the fractional organ volume receiving a dose Di and Equation E.4 is Sterling’s approximation for the equivalent square a is a tissue-specific parameter that describes the volume effect. of side c, for a rectangle of sides a and b. For a → −∞, gEUD approaches the minimum dose; thus negative Irregular Fields: For irregular fields, the equivalent square values of a are used for tumors. For a → +∞, gEUD approaches can be determined by separating the blocked areas into composite the maximum dose (serial organs). For a = 1, gEUD is equal to the rectangular fields, for which the equivalent square is known. Each arithmetic mean dose. For a = 0, gEUD is equal to the geometric composite equivalent square is then squared and summed, to form mean dose. gEUD is often used in plan evaluation and optimi- the ‘equivalent blocked area’. The square root of the difference zation because the same functional form can be applied to both between the equivalent blocked area and the open field area is the targets and OARs with a single parameter capturing (it is hoped) equivalent square of the blocked field. the dosimetric ‘essence’ of the biological response. For more complicated fields, Clarkson’s scatter integration Standard normal tissue complications probability (NTCP) method can be used, which is the most accurate, although the most models, such as the Lyman model, or tumour control probabil- labour intensive. This divides any irregular field into N equal nar- ity (TCP) models can be easily used only when an entire normal row sectors of circles with varying radii. The SAR is calculated organ receives a single homogeneous dose of radiation. This situ- as the average SAR of all N circles, and associated equivalent ation seldom applies in clinical radiotherapy. In clinical practice square size found. usually involving delivery of external beam radiotherapy (EBRT), Further Readings: British Institute of Radiology. 1996. each OAR receives a range of doses of ionising radiation. As a Central Axis Depth Dose Data for Use in Radiotherapy, British result, models have been developed that consider the inhomoge- Journal of Radiology, BJR Supplement 25, London, UK; neous dose distribution using so-called dose-volume reduction Clarkson, J. R. 1941. A note on depth doses in fields of irregular methods. EUD can be considered a dose volume reduction tech- shapes. Br. J. Radiol. 14:255. nique for the purposes of probability calculations. EUD can be used for biologically based plan evaluation, Equivalent tissue air ratio (ETAR) having the advantage of fewer model parameters, as compared (Radiotherapy) The equivalent tissue air ratio (ETAR) method is to standard NTCP/TCP models. This allows for more clinical a sophisticated 3D technique used to correct isodose data for tis- flexibility. sue heterogeneity corrections for use in treatment planning. Further Readings: Li, X. A. (Chair) et al. 2012. The use and It uses electron density information from 3D CT scans to cal- QA of biologically related models for treatment planning, report culate a weighted average of the electron densities, which is used of AAPM task group 166 of the therapy physics committee. Med. to predict the scattering effect of the surrounding structures. Phys. 39(3); Niemierko, A. 1997. Reporting and

analyzing dose Other methods for tissue heterogeneity corrections include the distributions: A concept of equivalent uniform dose. Med. Phys. ‘effective depth’ method, and the ‘Batho’ or ‘power law’ method. 24(1):103–110; Niemierko, A. 1999. A generalized concept of equiv- These are summarised in Report 24 from the International alent uniform dose (EUD). Med. Phys. 26:1101. Abstract WE-C2-9. Commission on Radiation Units and Measurements. Abbreviations: CT = Computed tomography and ETAR = Erect Equivalent tissue air ratio. (General) There are a series of terms used to describe the position Related Articles: Heterogeneity, Treatment planning system of an individual when undertaking different imaging examination. Further Readings: ICRU (International Commission on Erect: Standing or sitting up. For example, an erect chest x-ray. Radiation Units and Measurements). 1976. Determination of Related Article: Patient position ERM project 342 Ethics of radiation protection ERM project Erythema (General) Education for Radiation in Medicine (ERM) is an (Non-Ionising Radiation) Erythema is the redness of the skin international project, supported by the EU Tempus programme caused by increased blood flow in superficial blood vessels. (1995–1998). The ERM project (together with the EMERALD This occurs with UV induced damage, after UV exposure or project) was the result of the first International Conference on other skin injuries. There are different stages of erythema, rang- Medical Physics Education (Budapest, 1994). The partners in ing from a just perceptible redness to painful blistery rashes. the ERM project were organised in a consortium of universi- Related Articles: AORD, UV light hazard ties as follows: King’s College London, University of Florence, Further Readings: Coleman, A., F. Fedele, M. Khazova, Trinity College Dublin, Medical University Plovdiv, Technical P. Freeman and R. Sarkany. 2010. A survey of the optical haz- University, Plovdiv, University of Plovdiv. The project manager ards associated with hospital light sources with reference to the was C. Roberts and the coordinator was S. Tabakov. Partners’ Control of Artificial Optical Radiation at Work Regulations 2010. representatives included F. Milano, N. Sheahan, A. Djurdjev, K. J. Radiol. Prot. 30(3):469; ICNIRP. A closer look at the thresholds Velkova, A. Litchev, G. Stoilov and N. Balabanov. of thermal damage: Workshop report by an ICNIRP task group. The ERM project developed a complete MSc curriculum Health Phys. 111(3):300–306; ICNIRP. 2013. Guidelines on limits with full syllabi and lectures in English (20 textbooks with over of exposure to incoherent visible and infrared radiation. Health 3000 pages in total). The projects developed an Inter-University Phys. 105(1):74–91; ICNIRP. 2013. Guidelines on limits of expo- Medical Physics Educational Centre in Plovdiv, Bulgaria, thus sure to laser radiation of wavelengths between 180 nm and 1,000 providing a model for other small countries to develop their medi- μm. Health Phys. 105(3):271–295; ICNIRP. 2004. Guidelines on E cal physics university courses. The methodology of this project limits of exposure to ultraviolet radiation of wavelengths between was used as the basis of a later international project with EU 180 nm and 400 nm (Incoherent Optical Radiation). Health Phys. funding (1998–2000) developing MSc programmes in the Baltic 87(2):171–186; ICNIRP. 2000. Revision of the guidelines on lim- states (Estonia, Latvia, Lithuania). Fundamental elements of the its of exposure to laser radiation of wavelengths between 400 nm ERM project were used in support of the establishment of MSc and 1.4 μm. Health Phys. 79(4):431–440. programmes in medical physics in the Czech Republic, Malaysia, Thailand, Jamaica, Italy, France, Malta, UAE, Zimbabwe. The Escape peak ERM MSc programme in Plovdiv was the first one outside the UK (Nuclear Medicine) This is a peak in the energy spectrum of a NaI to receive IPEM accreditation (in 1998). (Tl). Following an interaction with an iodine atom in a NaI (Tl) Further Readings: Auzinsh, M., J. Spigulis, Y. Dekhtyar, crystal, a characteristic iodine K-x-ray (E = 30 keV) is emitted. I. Knets, A. Katashev, H. Hinrikus, K. Meigas, P. Kingisepp, For example, a photoelectric absorption of a 197Hg gamma ray (E A. Soosaar, A. Lukosevicius, A. Oberg, C. Roberts, S. Tabakov = 77.3 keV) followed by an escape from the detector by the char- and J. Lee. 2000. Baltic Biomedical Engineering and Physics acteristic K-x-ray results in a registered energy of 47.3 keV. The MSc Courses, Riga, Latvia, ISBN 9984 681 52 1; Tabakov. S., effect is most prominent when acquiring low energy gamma rays C. Roberts, F. Milano, N. Sheahan, V. Tabakova, K. Velkova, with a thin crystal because the K-x-ray is more likely to escape. A. Litchev, G. Stoilov, G. Petrova, G. Spassov, M. Yaneva, L. Also the relative distance between the two peaks gets smaller Michova, E. Milieva, S. Kostianev, N. Boyadjiev, G. Belev, N. when the gamma ray energy increases. Balabanov, M. Mitrkov, P. Trindev, I. Daskalov, D. Pressianov Further Reading: Cherry, S. R., J. A. Sorenson and M. E. and M. Nencev. 1999. Inter-university education centre and MSc Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, course in medical radiation physics and engineering. J. Med. Philadelphia, PA, p. 152. Biol. Eng. Comput. 37(s1):153–154; Tabakov, S. and V. Tabakova. 2015. The Pioneering of e-Learning in Medical Physics, Valonius Escargot curves Press, London, available free at: www .emerald2 .eu /mep _15 .html (Radiotherapy, Brachytherapy) Related Articles: EMERALD project, EMERALD II project, Paris Dosimetry System: For the Paris system dose calcula- EMIT project, e-Learning tions, dose rates in water at the central plane of a linear source are needed, where oblique filtration and attenuation and scattering in water are taken into account. Dose rate information for 192Ir wires Ernst angle is available for instance in Dutreix et al. (1982) and Pierquin and (Magnetic Resonance) The Ernst angle is the flip angle which Marinello (1997). The information is given both graphically and maximises signal in a spoiled gradient echo sequence for a given in tables. In the graphical presentation, the ‘escargot curve’, the TR time and a given tissue T1. The Ernst angle is given by central plane dose rate information is summarised in one single graph, showing the dose rate at varying distances for wires of unit qernst = cos-1 (e-TR/T1 ) linear source strength and different lengths. Escargot means snail in French and the graphs have a shape that resembles a snail’s shell. An optimum flip angle exists because two competing factors Related Articles: Paris system, Cross line curves are at work in the relationship between signal amplitude and flip Further Readings: Dutreix, A., G. Marinello and A. angle in a spoiled gradient echo sequence. A smaller flip angle Wambersie. 1982. Dosimétrie en Curiethérapie, Masson, Paris, causes longitudinal magnetisation to settle into a stronger steady France; Pierquin, B. and G. Marinello. 1997. A Practical Manual state. A stronger steady-state longitudinal magnetisation favours a of Brachytherapy, Medical Physics Publishing, Madison, WI; stronger signal as more magnetisation is available to flip onto the Schlienger, M. et al. 1970. Contrôle dosimétrique en brachycu- transverse plane. However, against this, a smaller flip angle flips riethérapie par les isodoses ‘escargot’. Acta Radiol. 9:282–288. less of the longitudinal magnetisation onto the transverse plane in each TR period, making less available for signal formation. The Ethics of radiation protection Ernst angle (named after R. Ernst) is the balance point between (Radiation Protection) The Ethical Foundations of the System these competing factors, where signal is maximised. of Radiological Protection are set out in the International EURATOM Treaty 343 European Guidelines on Medical Physics Expert project Commission for Radiological Protection Report 138. The report Therefore the radiation protection unit, based in Luxembourg, defines an ‘ethical value set’ derived from classical ethical values deals with the following topics: (1) exposure of public; (2) occu- and updated to reflect the uses of ionising radiation that lead to pational exposures; (3) emergency preparedness and response; occupational, medical and public exposure. (4) natural radiations; (5) medical exposures; (6) environmental The core ethical values can be summarised as follows: monitoring and assessment and (7) education, information and • Beneficence/Non-Maleficence training. A different section deals with the effects of non-ionising Beneficence/non-maleficence is the ethical basis for radiations. the radiation protection principle of justification – that In order to reach the objectives, it is necessary: (1) to propose the exposure should ‘do good’ – i.e. be beneficial – and and to implement community legislations; (2) to check the legal that it should do no harm. and operational implementation of community legislations; (3) to • Prudence prepare basic safety standards; (4) to verify that member states Prudence is related to the precautionary principle – perform their duties regarding obligatory monitoring of environ- that urges caution before making the decision to expose mental radioactivity; (5) to provide a system of rapid information a person to ionising radiation, and that it should be exchange in case of nuclear accidents; (6) to ensure implemen- evidence-based. tation of maximum permitted levels in food and (7) to discuss • Justice and Dignity radiation protection issues in meetings with the participation of Justice and dignity are ethical values for ensuring independent experts. that all persons exposed are treated equally and fairly Hyperlink: EURATOM: http://www .euratom .org and ensuring that those exposed give informed consent. E EUD (Equivalent uniform dose) These four basic ethical values are supported by a further number (Radiotherapy) See Equivalent uniform dose (EUD) of procedural values: • Accountability The ethical value of accountability ensures that there European Guidelines on Medical Physics Expert project is always someone recognised as taking responsibility (General) The purpose of the EU-funded European Guidelines for the decision to expose a person to ionising radiation. on Medical Physics Expert (MPE) project was to provide for • Transparency and Inclusiveness improved implementation of the provisions of EU Directive Transparency implies that all decisions are made in 2013/59/EURATOM related to the role of the MPE and to facili- an open manner that can be scrutinised by stakehold- tate the harmonisation of the education and training of medi- ers (management, staff, regulatory authorities, patients cal physicists to MPE level among the EU member states. The and the public). This is supported by the ethical value of document contains recommendations on harmonising education, inclusiveness – ensuring that all stakeholders who may training and recognition requirements for MPEs in the European be affected by the decision to expose persons to ionis- Union within the existing EU legislative framework. It makes ing radiation have been informed and involved where recommendations for the most appropriate education and train- appropriate in that decision. ing structure, based on the European Higher Education Area and on the European Qualifications Framework for Lifelong Learning Related Articles: Informed consent, Justification, to achieve the required professional competences. It proposes Optimisation, Limitation, Stakeholder involvement detailed education and training learning outcomes for MPEs in terms of knowledge, skills and competences. The document also EURATOM Treaty contains recommendations on the minimum MPE staffing levels (Radiation Protection) The European Atomic Energy Community necessary to ensure adequate radiation protection of patients. It was established with a Treaty in 1957. This is one of the found- also includes recommendations on radiation protection of staff ing treaties establishing the European Union (EU). The Treaty, when impacting medical exposure. drafted in 1950, addressed mainly the topics considered impor- The project was coordinated by: tant at that time, in particular regarding the field of nuclear power. Department of Radiology, Facultad de Medicina, Universidad Under the provision of the EURATOM Treaty, the European Complutense de Madrid, Spain (Eduardo Guibelalde .egc @med . Commission acquired the status of a supranational regulatory uc m .es) authority in the following fields: radiation protection, supply of nuclear fissile materials and nuclear safeguards. Of these the former is relevant to the operation of all facilities including also The other members of the consortium were: medical applications and hospitals. • European Federation of Organisations for Medical The EURATOM Treaty makes little mention of criteria and Physics (EFOMP) norms to be respected during design or operation of facilities. • Institute of Physics and Engineering in Medicine, UK Therefore these kinds of regulatory activities have been devel- • Department of Physics, ‘Enrico Fermi’, University of oped under the responsibility of national authorities. International Pisa, Italy organisations and in particular the International Atomic Energy • German Society of Medical Physics (DGMP), Germany Agency (IAEA), have been able to achieve a certain level of stan- • North East Strategic Health Authority, Yorkshire and dardisation. After years of work and several international conven- the Humber, UK tions a ‘Culture’ of best

practice has been established among the • Quality Assurance Reference Centre, UK member states and therefore also the EU countries. The division between safety and health protection of the workers and public Hyperlinks: The European Guidelines on the MPE was pub- appears somehow ‘arbitrary’; there is in fact an effort towards a lished by the European Commission as Radiation Protection more comprehensive approach. Series 174 and can be downloaded from here: https :/ /op .euro pa .eu The protection of the health of the workers and public from /en /p ublic ation -deta il/-/ publi catio n /b82 ed768 -4c50 -4c9a -a789 the harmful effect of ionising radiations is the overall objective. -98a3 b0 df5 391 /l angua ge -en EUTEMPE Project 344 Event type in PET EUTEMPE Project (European Training and (COMAC-BME EC project, 1995), indications have been given on Education for Medical Physics Experts Project) basic parameters necessary to establish MRS/MRI image qual- (General) The aim of the EU funded project EUTEMPE-RX was ity. Specific cylindrical phantoms or test objects have been made to produce a set of education and training modules for helping by many manufacturers to allow their measurement. Phantoms medical physicists attain medical physics expert (MPE) status in are formed by slices and contain objects that allow you to assess diagnostic and interventional radiology. The modules are ongo- uniformity, ghosting, image signal-to-noise ratio and uniformity ing. It is planned to set up similar consortia to produce paral- of image signal-to-noise ratio, geometric distortion, slice profile lel sets of modules for the other two main areas of specialisation and slice width, slice warp, slice position, resolution, modula- of medical physics, i.e. radiation oncology (EUTEMPE-RO) and tion transfer function, T1 and T2 precision and accuracy, image nuclear medicine (EUTEMPE-NM). To this effect the consor- contrast. tium is now referred to as EUTEMPE-NET. Further Readings: IV. 1988. Protocols and test objects for the The original EUTEMPE-RX project was coordinated by: assessment of MRI equipment. Magn. Reson. Imaging 6:195–199; Katholieke Universiteit Leuven, Belgium (Hilde Bosmans, hilde Lerski R, Certaines J. 1993. Performance assessment and qual- .bosmans @uzleuven .be) ity control in MRI by Eurospin test objects and protocols. Magn. The other members of the consortium were: Reson. Imaging 11(6):817–33. • European Federation of Organisations for Medical Physics (EFOMP) Event-related design • Servicio Madrileno de Salud, Spain (Magnetic Resonance) During a fMRI study, the subject under- E • Università degli Studi di Pavia, Italy takes a series of tasks known as a paradigm. These tasks may be • Universitat Politecnica de Catalunya, Spain arranged either in a block design (also known as a boxcar), or an • Università degli Studi di Ferrara, Italy event-related design. • Technical University of Varna, Bulgaria For an event-related design, stimuli (events) are presented not • Royal Surrey County Hospital NHS Foundation Trust, UK in ordered epochs, but briefly and randomly, so that the partici- • Centre Hospitalier, Universitaire,Vaudois, Switzerland pant cannot predict if or when they will occur. Such experimental • Stichting Landelijk, Referentie Centrum voor design allows the detection and analysis of the haemodynamic Bevolkingsonderzoek, Netherlands response function associated with individual events (in contrast • Panepistimio Kritis, Greece with block designs, where there is simply a temporal integration • Azienda Sanitaria Universitaria Integrata di Udine, Italy of signal). By randomly mixing events of different types, it is pos- • Stadisches Klinikum Braaunschweig GMBH, Germany sible to ensure that the response to any one event is not systemati- • Technische Hochschule Mittelhessen, Germany cally influenced by previous events (Figure E.48). Developments in event-related design have largely been Hyperlinks: http://eutempe -net .eu/ fuelled by advances in the speed with which fMRI data can be acquired. Statistically, event related design is not as powerful as Eurospin test objects block design. (Magnetic Resonance) In the frame of the European Related Articles: fMRI (functional magnetic resonance imag- Concerted Action ‘Tissue Characterization by MRS and MRI’ ing), Block design, Haemodynamic response function Further Reading: Amaro, E. and J. Barker. 2006. Study design in fMRI: Basic principles. Brain and Cognition 60(3), 220–232. T T T T T Event-related design Event type in PET (Nuclear Medicine) The annihilation coincidence detection (ACD) localises events along a line of response (LOR) when two coinci- dences are recorded in two opposite detectors within a specified FIGURE E.48 Event-related design where stimuli/events (top) are coincidence timing window. If the detected photons originate denoted T. The corresponding BOLD responses are shown (bottom). Note from the same annihilation and have travelled without any scat- that the response for each stimulus can be detected and analysed in detail. tering events, the event is referred to as a true coincidence (see (From Amaro, E. and Barker J., Brain Cognit., 60(3), 220, 2006.) Figure E.49a). The true coincidences are a good representation (a) (b) (c) FIGURE E.49 (a) True coincidence, (b) scattered coincidence and (c) random coincidence. Event types in a scintillation camera 345 Event types in a scintillation camera of the actual activity distribution. All coincidences recorded by for random coincidences) just outside the object. This profile is the coincidence processor may not be true coincidences. Other assumed to represent the distribution of scattered photons and coincidence events can also occur within the resolving time of the the distribution is extrapolated over the entire projection using a detector system. Two examples are shown in Figure E.49b and c. cosine or Gaussian distribution. As seen in Figure E.49c, a random coincidence occurs when Related Articles: PET, Beta decay two photons with unrelated annihilation points are detected Further Reading: Cherry, S. R., J. A. Sorenson and M. E. within the coincidence timing window. The count rate for random Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, coincidences for a detector pair is given by Philadelphia, PA, pp. 340–342. R Event types in a scintillation camera random = CTW ´ Rsingle,1 ´ Rsingle,2 (E.6) (Nuclear Medicine) There are four basic event types that can be registered in a gamma camera system. Only one of these event where types provides correct positional information. The other event CTW is the coincidence timing window types cause degradation in spatial resolution (contrast) and errors Rsingle,1 and Rsingle,2 are the counting rates for the respective in activity quantification. The first event type is called valid events. single channel detector A valid event occurs when a photon is emitted near-parallel to the collimator holes, passes through the collimator and depos- With greater administrated activity the quotient between ran- its all its energy in the crystal with one photoelectric interaction. dom and true coincidences increases. As it follows from the pre- This is the event type that provides correct spatial information. E ceding equation the count rate for random coincidences increases Camera systems today are designed to increase the fraction of as the square of the activity present while the amount of true valid events by using crystal material with high atomic number, coincidences increases linearly. From the equation it also fol- specially designed collimators with regard to photon energy and lows that the amount of random coincidences is proportional to energy window discrimination. the coincidence timing window. There is a lower limit for this The second event type is the detector scatter event. These window due to limitations in the electronics and time of flight events occur when a photon is emitted near-parallel to the col- considerations. The typical quotient for random to true coinci- limator holes and Compton scattering in the crystal occurs. The dences in clinical PET scanners is 0.1–0.2 (brain imaging) and scattered photon can either interact in a different location in the sometimes >1 for certain applications such as whole body imag- crystal or escape. In the latter case it is possible to discriminate ing. Random coincidences appear fairly uniform across the field events by analysing the pulse height, which is proportional to the of view (FOV) and cause a loss of image contrast. It is also a energy deposited. If the energy deposited is not enough to reach source of error when quantifying the activity within the patient. the energy window the event is discarded. But if the scattered One way to correct for random coincidences is to delay the coin- photon interacts in a different part of the crystal then the total cidence time window. The delay should be much greater than the amount of deposit energy may be within the energy window. width of the window itself (e.g. 128 ns where width of window The location of the event will therefore be placed somewhere in- is 6 ns). The coincidences detected in this way are all random between the two interactions. At high count rates, pile up can lead and will be approximately equal to the number of random coinci- to a very prominent image artefact known as the pile up effect dences detected in the prompt coincidence window. A correction (separate article). Another type of detector scatter event occurs can be made by subtracting the delayed coincidences. A third when photons scatter in the collimator before being registered. category of events is when one or both photons are scattered Such an event can be discriminated using an energy window. But before detection and as a result the LOR is misplaced. These if the energy resolution of the crystal is low some of these events coincidences are called scattered coincidences (Figure E.49b). might be registered as true events. These events will cause a loss The magnitude of the displacement is determined by the scatter of contrast. angle and the location of the scatter event. The scatter event will The third event is the object scatter event. Such an event most likely occur inside the patient but can also occur within occurs when the photons are not emitted towards the collima- components inside the scanner. The fraction of scattered events tor but scatter inside the object, pass through the collimator and can in some cases (abdominal imaging) be as high as 60%–70%. deposit all energy at a single location. Such an event can be mis- This high fraction has three different causes. The first cause is located by several centimetres. The fraction of scattered events that only one of the two scattering photons must scatter to pro- can be reduced by the use of an energy window, i.e. all events that duce a misplaced LoR. The second cause is that Compton scatter do not register an energy deposition within the energy window are is the dominating interaction in scintillators at 511 keV and some rejected. But if the scatter angle is less than 45° the loss of energy incident photons might only deposit a small amount of energy. is minimal and the problem still remains. Object scattered events As a result the pulse height analysis window must be widened result in a low-spatial-frequency background that decreases the to include these interactions. The final cause is the low energy image contrast. Another source for scatter events is the collima- resolution in the LSO and BGO scintillators because of the low tor, i.e. when a photon Compton scatters in the collimator before light output. It is therefore customary to use a wide pulse height being registered. analyser window. It is impossible to separate scattered events The fourth event is septal penetration. A photon emitted that originate in the body from those that originate inside the towards the collimator can penetrate the septal walls and inter- crystal and therefore the scatter method used in SPECT, e.g. a act in the crystal. This will cause a blurriness in the image since double-energy window is less successful. Today there are two the system will assume that the event originated from a location main methods to deal with scatter in PET. The first one uses a perpendicular to the collimator face. This effect is more promi- transmission image and makes computational estimations of the nent when using high energy gamma-emitters. A way to mini- scatter distribution for each projection. The second approach is mise the number of septal penetration events is to use a collimator to look at the scatter distribution in a projection profile (corrected designed for the specific energy. Evidence-based 346 Exposure Events that are combinations of these simple events can also Dexit could be derived from measurements with a detector placed occur but are not as frequent. at dmax from the exit

surface where dmax is the build-up depth. The Related Articles: Parallel-hole collimator, Pile up effect detector used in the exit dose determination should have a build- Further Reading: Cherry, S. R., J. A. Sorenson and M. E. up thickness around its sensitive volume to ensure complete elec- Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, tron backscatter. The same irradiation conditions must be ensured Philadelphia, PA, pp. 222–223. in the case of the determination in a phantom of the ratio between the dose to the detector and the patient exit dose. Evidence-based (General) Any practice that relies on scientific evidence for guid- Exit window ance and decision-making is called ‘evidence-based’. Practices (Diagnostic Radiology) The term exit window is used to describe that are not evidence-based may rely on tradition, intuition and the part of the x-ray tube glass envelope through which the useful non-systematic experience. Evidence-based practice is the inte- beam of x-rays exits the tube. At this place, the glass of the tube gration of best research evidence with professional expertise and envelope (usually borosilicate glass) is made thinner in order to is outcome centred. When applied by practitioners this will ulti- reduce the absorption of the useful beam. When the tube is placed mately lead to optimal results. in the tube housing the exit window is adjusted to face the open- ing of the housing. In x-ray tubes with metal housing the exit win- Excess risk dow is made from metal with low absorption – usually beryllium (Radiation Protection) Excess risk is used to compare the risk (Z = 4) – this is typical for mammographic x-ray tubes (1 mm is a E in two different groups of people. For example, one group common thickness). exposed to radiation for medical purposes, and the other group Related Articles: Glass envelope, X-ray tube not exposed. In this example, the groups may be compared to one another Expert systems epidemiologically to see if belonging to that group exposed to the (Diagnostic Radiology) Expert systems are artificial intelligence medical radiation increases their risk of developing certain dis- programs that attempt to reproduce or emulate the performance of eases, particularly cancer. Any increase in the number of cancers a human expert in a particular specialist problem solving task. In detected in the exposed group, over and above that in the unex- order to simulate the performance of an expert a knowledge base posed group may be termed the excess risk. has to be formed as the system learns from the performance of a Excess risk should be analysed in comparison to total or abso- specialist or subject matter expert. The expert system is then com- lute risk, and the relative risk compared to other factors. pared to a human expert to verify its performance. An example Related Articles: Absolute risk, Relative risk of an expert system in radiology is in computer aided detection as used to identify suspicious lesions in digital mammography or for Excitation detecting lung nodules in pulmonary CT. (Magnetic Resonance) In physics, excitation is the elevation of an energy level over an arbitrary ground energy state. The lifetime Exposure of an excited system is usually short and determined by relax- (Radiation Protection) The quantity of X or gamma radiation ation processes. In MR an excitation is obtained by transferring measured in terms of the ionisation produced in air is known as energy into the spinning nuclei using a radio frequency (RF) exposure. pulse. The length and amplitude of the excitation RF-pulse is Exposure is the quantity of X or gamma radiation measured chosen in a way (flip-angle) to produce net transverse magnetisa- as the sum of the electric charges (Q) on all the ions of one sign tion, which is measured afterwards. In MRI, after excitation data (+ve or −ve) liberated and collected in air when all the secondary in k-space is measured. Furthermore, a number of excitations and electrons liberated by the photons in a small mass (dm) of air are subsequent measurements with the same parameter settings are completely stopped in air: used to average the data and sometimes called number of excita- tions (NEX). dQ Related Articles: Transverse magnetisation, Relaxation, Flip X = dm angle The historical unit of exposure is the Roentgen (R), while the SI unit is Coulomb per kg: Exit dose (Radiotherapy) The exit dose Dexit is the absorbed dose delivered to a surface where the central axis of radiation emerges from the 1R = 2.58´10-4 C/kg patient. At the patient exit side there is a lack of backscatter radia- tion because of a reduction in either the number of backscatter Exposure does not include ionisation due to secondary photons photons or secondary backscatter electrons. These effects result (e.g. Bremsstrahlung), therefore exposure can be considered to be in a build down layer near the exit surface of the patient. The lack the ionisation equivalent of air kerma. of backscatter photons is dependent on the field size and increases This is an important definition – ionisation chambers are used with the patient thickness and it is higher with lower photon beam for many applications for the actual physical measurement of energy. The lack of photon backscatter influences a large region. radiation output, e.g. from an x-ray or radiotherapy unit. The sub- The lack of secondary backscatter electrons influences the last sequent conversion of that measurement into radiation dose units millimetres of the patient tissues in front of the exit surface. The (e.g. absorbed dose in air, air kerma, etc.) is a purely mathematical position at which the exit dose should be defined must be carefully exercise. evaluated. Even if the build-up and build-down thicknesses are Related Articles: Roentgen (R), Absorbed dose, Air kerma, different they can be assumed as equal for many measurements. Bremsstrahlung Exposure counter 347 Exposure point Exposure counter • B(λ): A blue light action spectrum, which takes into (Diagnostic Radiology) A device (or function) within an x-ray account the photochemical interactions between visible machine to record the number of exposures produced. light (primarily around the blue region) • R(λ): A thermal action spectrum which takes into account the heating hazards on both the eye and skin of Exposure index visible and infrared radiation. (Diagnostic Radiology) The exposure index (EI) in digital radiog- raphy has been used to indicate the relative speed and sensitivity of a digital receptor to incident x-rays. Exposure index is sup- The spectra are tabulated into the directive. posed to provide feedback to radiographers regarding the proper The ICNIRP originally published guidelines on exposure lim- radiographic techniques used in a specific exam that obtain the its for incoherent visible and infrared (1997) and UV radiation appropriate image quality with the lowest dose to the patient. EIs (1996), and for laser radiation of wavelengths between 1000 µm were entirely manufacturer-specific and varied greatly in termi- and 180 nm (1996). These limits have been updated and reviewed nology, mathematical forms and calibration conditions. Recently since they were first published. In 2000 a separate set of guidelines the International Electrotechnical Commission (IEC) developed was published regarding limits of exposure to laser wavelengths a standardised terminology for EI. The standardised EI should be in the retinal hazard region (400 nm–1.4 µm). The introduction of consistent between manufacturers and reflect the radiation inci- the EU Directive on health and safety requirements for workers dent on the detector and is based upon image noise levels. It is exposed to artificial optical radiation in 2006 (2006/25/EC), saw important to stress that EI does not reflect the radiation dose to the adoption of the ICNIRP limits into European law. the patient. Related Articles: Action spectra, AORD, Effective exposure, E ICNIRP, Eye, Skin Further Readings: Coleman, A., F. Fedele, M. Khazova, Exposure limit values P. Freeman and R. Sarkany. 2010. A survey of the optical haz- (Non-Ionising Radiation) These are the maximum exposure ards associated with hospital light sources with reference to the established by ICNIRP and indicated in the AORD 2006 which Control of Artificial Optical Radiation at Work Regulations 2010. no significant lasting biological effect has been seen. In the case J. Radiol. Prot. 30(3):469; ICNIRP. A closer look at the thresholds of optical radiation the two organs affected are the eyes and the of thermal damage: Workshop report by an ICNIRP task group. skin. Health Phys. 111(3):300–306; ICNIRP. 2013. Guidelines on limits Exposures are expressed as radiant exposures for the skin of exposure to incoherent visible and infrared radiation. Health (Jm−2) and radiances for the eye (Jm−2sr−1). Radiant exposures Phys. 105(1):74–91; ICNIRP. 2013. Guidelines on limits of expo- refer to the light received on the skin per metre square, whilst sure to laser radiation of wavelengths between 180 nm and 1,000 radiances refer to the light that has been transmitted through a μm. Health Phys. 105(3):271–295; ICNIRP. 2004. Guidelines on unitary solid angle. limits of exposure to ultraviolet radiation of wavelengths between The exposure is estimated taking into account the biological 180 nm and 400 nm (Incoherent Optical Radiation). Health Phys. interaction of light according to its energy (i.e. its wavelength), by 87(2):171–186; ICNIRP. 2000. Revision of the guidelines on lim- weighting the exposure by photo biological action spectra. its of exposure to laser radiation of wavelengths between 400 nm Only a subset of the limits would usually apply to non-laser and 1.4 μm. Health Phys. 79(4):431–440. hospital sources (Table E.4). As one can see from the table, there are three key action Exposure point spectra: (Diagnostic Radiology) Exposure point (or exposure number, or L number) is an old measure used for adjusting the param- • S(λ): An ultraviolet action spectrum, which is based on eters of the x-ray exposure used by radiographers in some coun- the light-skin interactions tries. This measure is analogous to similar measure used in film TABLE E.4 Exposure Limits that Generally Apply to Non-Laser Hospital Artificial Optical Sources Limit Spectral Range Action Index Hazard Physical Quantity (nm) Spectrum ELV a Actinic hazard (Heff), Effective radiant 180–400 S(λ) 30 Jm−2 Daily value 8 hours exposure b Near-UV hazard Radiant exposure 315–400 104 Jm−2 Daily value 8 hours (HUVA), d Retinal blue-light Effective radiance 300–700 B(λ) 100 Wm−2sr−1 for t > 10,000s hazard (LB) For α ≥ 11 mrad t > 10,000s g Retinal thermal Effective radiance 380–1400 R(λ) 2.8 107/Cα Wm−2sr−1 for t > 10s where hazard (LR) Cα = α for t > 10s 1.7 ≤ α ≤ 100 mrad Cα = 100 for α > 100 mrad Exposure rate 348 Extended field of view (FOV) photography. As per the exposure point system, introduced by external ionising radiation and/or internal contamination. The Klaasen, one exposure point is equal to the necessary change of evaluation should be made preventively, on the basis of the work- x-ray exposure (dose), which will produce exactly the same dark- load and of the work organisation, by the radiation protection offi- ening of the x-ray film, if the object thickness changes with 1 cer/qualified expert. The exposure time allowed, for each worker, cm. Based on exposure points radiographers had built exposure for each working condition should be as precise as possible, in tables, with the help of which they can easily calculate the optimal order to take into account the characteristics and the risk related exposure for a radiograph by simply measuring the thickness of to the various working conditions. The aim is to be sure that the the patient. The introduction of AEC (automatic exposure control) occupational workers are operating, with sufficient confidence has replaced this system. and safety, within the occupational dose limits. Standard opera- tional practices shall be indicated in order to minimise the risk Exposure rate and optimise the procedures. Training of personnel and routine (Diagnostic Radiology) The rate at which radiation is delivered verifications of the working conditions and related risks, for the to a specified location (as mR/s). For example, the exposure rate workers, are an essential part of any safety assessment programs. produced by a fluoroscope at the table surface. Further Reading: 1996. International Basic Safety Standards for Protection against Radiation and for the Safety of Radiation Exposure switch Sources, Safety Series No. 115, International Atomic Energy (Diagnostic Radiology) Normally the exposure switch of an x-ray Agency, Vienna, Austria. radiograph system initiates the exposure in two stages. The first E stage (preparation) heats the cathode to

the requested temperature Extended source (optical) and rotates the anode to the requested rotational speed. The sec- (Non-Ionising Radiation) An optical source is defined as ond stage (exposure) applies the high voltage (kV) and initiates extended, from an optical point of view, when its dimension and the exposure. Due to this reason the exposure switch normally has distance from the viewer is such that the angle subtended to the two buttons one into the other. Releasing the switch usually does eye is larger than 1.5 mrad, which corresponds to a retinal spot not affect the exposure, as its end is determined by the system size of about 25 µm. timer. Most sources used in healthcare are extended sources. Related Article: High voltage generator Related Articles: ICNIRP, Exposure limit values Further Readings: BS EN 62471. 2008. Photobiological safety Exposure table of lamps and lamp systems; Council Directive 2006/25/EC on the (Diagnostic Radiology) An exposure table, also known as a tech- minimum health and safety requirements regarding the exposure nique chart, is a table relating x-ray exposure factors such as KV, of workers to risks arising from physical agents (artificial opti- MAS, FRD, etc. to patient body characteristics and is used to cal radiation) (19th individual Directive within the meaning of set up radiographic procedures. Exposure tables were made for Article 16(1) of Directive 89/391/EEC) [2006] OJ L 114; ICNIRP. all x-ray equipment, related x-ray films and anatomical regions 2013.Guidelines on limits of exposure to incoherent visible and to be imaged. Initially many AEC (automatic exposure control) infrared radiation. Health Phys. 105(1):74–91; A Non-Binding systems used existing exposure tables to determine the correct Guide to the Artificial Optical Radiation Directive 2006/25/EC, x-ray parameters. The introduction of AEC with dose detectors Radiation Protection Division, Health Protection Agency. https has replaced this system. :/ /os ha .eu ropa. eu /en /legi slati on /gu ideli nes /n on -bi nding -guid e -to- Related Article: Automatic exposure control good- pract ice -f or -im pleme nting -dire ctive -2006 -25 -e c -201 aarti ficia l -opt ical- radia tion2 019. Last accessed January 2020. Exposure time (Diagnostic Radiology) Exposure time is the duration of the time Extended field of view (FOV) the x-radiation during a radiographic procedure (usually mea- (Ultrasound) In current commercial ultrasound scanners, the sured in ms). It is either manually set by the radiographer or con- FOV of the displayed image is dictated by the design of the trans- trolled by the automatic exposure control (AEC) function of the ducer and its operation. For linear arrays, the width of the image equipment based on measurement of radiation passing through is determined by the length of the array; for curvilinear arrays the patient and reaching the receptor. AEC is sometimes known – by the length of the array and its radius of curvature – and in as phototiming. phased arrays – by the length of the array and sector angle. Historically, extended-FOV images were available by combin- Exposure time ing images from an array of separate transducers or by storing (Radiotherapy) This is the time for which the radiation field is several images from different transducer positions and using them applied. Sometimes it is called beam on time. for a composite image. This term is used for radionuclide sources (Cobalt unit, Location of the image position is key to this and was achieved brachytherapy). in these systems by using multiple fixed transducers or by tracking For external beam radiotherapy (linear accelerators, x-ray the probe position. Optical and magnetic tracking of ultrasound orthovoltage and superficial machines), this concept has largely probe position is still used for image registration in research been replaced by monitor units. In other words, a number of moni- applications, for example in correlating ultrasound images with tor units is set up on the treatment machine instead of exposure 3-D MRI or CT neuro-imaging. time, in order to control the delivered dose. With increased computing power, several manufacturers offer Related Articles: Beam on time, Monitor unit extended-FOV imaging using the image itself as a reference and comparing features in subsequent images with the original Exposure time limit (Figure E.50). By estimating displacement of the images and the (Radiation Protection) Exposure time limits must be evaluated velocity of the probe movement, extended views of tissue along in the case of occupational working conditions with exposure to the axis of the transducer can be constructed (Figure E.51). Extended SSD treatment 349 E xternal beam irradiation Extended SSD treatment cylinder, that can move the restrictor closer to the radiographed (Radiotherapy) The term extended SSD treatment can be used in area without change in the focus to image distance. See article on two situations. In one the SSD is set to a value beyond the nomi- Beam restrictor. nal isocentre position of SSD = 100 cm to allow a larger field Related Articles: Beam restrictor, Cone size to be used. The other situation is one where surface obliquity prevents a standard offset being used with an electron applicator, Extent and so a larger SSD value must be used. For these situations it (Nuclear Medicine) The polar map image can be used to deter- is important that the virtual source position is known and so the mine the size and location of defects in a myocardial perfusion effective SSD can be found. scan (Figure E.52). The extent of a myocardial perfusion abnor- Abbreviation: SSD = Source to surface distance. mality is expressed as the number of pixels which fall below Related Articles: Effective source point, Virtual source posi- the normal limit as a percentage of the total number of pixels of tion, Source-to-skin distance (SSD), Electron applicator, Electron the myocardium. Extent and severity are two important param- oblique incidence, Effective SSD eters for assessment and management of coronary artery disease (CAD). Extension cylinder cone Related Article: Severity (Diagnostic Radiology) A simple beam restrictor – a metal extender, similar to cylinder cone, but with an additional outer External beam irradiation (Nuclear Medicine) Cyclotrons are used to produce radionuclides by accelerating charged particles on to a target. E Features recognized in each region are One way to irradiate the target is to insert it directly into the a common reference. particle beam path. This technique is called internal beam irra- e movement of the diation. Using this technique allows most of the accelerated par- transducer can be ticles to hit the target. determined from the The second alternative is to extract the particle beam from image itself. e images and their positional the cyclotron and direct it on to an external target using either information are used to a stripping foil or an electrostatic deflector. This is called exter- construct an extended field nal beam irradiation. In general, external beam radiation is the of view preferred method. One of the disadvantages with external beam radiation is the low extraction efficiency. As much as 30% of the beam is not extracted, especially for cyclotrons using positively charged particles (positive ion cyclotron). Here electrostatic deflectors are used. The non-extracted beam irradiates and acti- vates surrounding septa which leads to an unwanted radiation Region 1 Region 2 dose to cyclotron personnel. In modern negative ion cyclotrons, however, full extraction (close to 100%) is achieved with the use FIGURE E.50 Image-based extended FOV. of a stripping foil. Direction of movement Successive images combined – relative position based on common features to calculate movement Final extended field-of-view image FIGURE E.51 Extended-FOV image of a leg with oedema based on movement along the axis of a linear array. External beam therapy 350 Extinction cross section Photon External Radiotherapy: Photon external beams are mostly used in external beam radiotherapy. They are all charac- terised by the same physical parameters, but fall into various cate- gories depending on their origin, means of production and energy. There are two origins of photon beams: gamma rays, which origi- nate from radioactive nuclei, and x-rays, which originate in a tar- get bombarded with energetic electrons. The x-rays from a target consist of Bremsstrahlung photons (Bremsstrahlung x-rays) and characteristic photons (characteristic x-rays). X-rays are produced either in an x-ray tube (superficial or orthovoltage x-rays) or in a linear accelerator (megavoltage x-rays). External photon beam radiotherapy is usually carried out with more than one radiation beam in order to achieve a uniform dose distribution inside the target volume and as low as possible a dose in healthy tissues surrounding the target. Modern photon beam radiotherapy is carried out with a variety of beam energies and field sizes under one of two set-up conven- tions: a constant source to surface distance (SSD) for all beams or E constant source to axis distance (SAD) for all beams. In an SSD set-up, the distance from the source to the surface of the patient is kept constant for all beams and must be set up for every beam during irradiation, while for an SAD set-up the centre of the target volume (inside the patient) is placed at the machine isocentre once before the irradiation. Clinical photon beam energies range from low (30–80 kVp) used for treating skin cancer and superficial structures, through medium (100–300 kVp), to megavoltage energies (60Co to 25 MV) used to treat deep-seated tumours (e.g. bladder, bowel, prostate, lung, brain). There is no sharp division between the various volt- age ranges, and they can vary slightly in different documents. Field sizes range from small circular fields used in radiosur- gery, through standard rectangular and irregular fields, to very large fields used for total body irradiation (TBI). External beam radiotherapy is usually carried out in fractions. FIGURE E.52 Output of processing myocardial -perfusion abnormality. The total prescribed dose is divided into fractions (normal frac- tionation is one fraction per day, 5 days a week, 2 Gy per fraction). Related Articles: Cyclotron, Internal beam irradiation, Abbreviations: SAD = Source to axis distance, SSD = Source Stripping foil to surface distance and TBI = Total body irradiation. Further Reading: Cherry, S. R., J. A. Sorenson and M. E. Related Articles: Deep therapy, Linear accelerator, Conformal Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, radiotherapy, Photon beam, Electron beam, Heavy particle Philadelphia, PA, pp. 50–51. beams, Stereotactic radiosurgery, Radiotherapy, Intensity mod- ulated radiation therapy, Cobalt unit, Orthovoltage, Isocentre, External beam therapy Fractionation (Radiotherapy) Further Reading: Podgorsak, E. B. 2005. Radiation Oncology Vocabulary in Radiotherapy: Radiotherapy can be divided Physics: A Handbook for Teachers and Students, International into two types of therapy, teletherapy and brachytherapy. Atomic Energy Agency, Vienna, Austria. In teletherapy (from the Greek word tele which means ‘far off’), the sources are placed far from ‘the tumour’ (more than 5 cm). External photoelectric effect In brachytherapy (from the Greek word brachy which means (Radiation Protection) An interaction between photonic non-ion- ‘short’), the sources are placed close to or inside ‘the tumour’ ising radiation incident on the surface of a material (almost invari- (less than 5 cm). ably a metal) in which the absorption of a photon in the material External beam therapy is in everyday speech said to be identi- results in the excitation of an electron such that it is liberated from cal with teletherapy. It is to be noted though, that there are exter- the surface of the material. nal beam techniques with very short source-skin distances, such Related Articles: Photoelectric effect, Internal photoelectric as contact treatment with a contact x-ray machine, using a source- effect skin distance of 4 cm. External Beam Therapy Characteristics in General: In Extinction cross section external beam radiotherapy the radiation source is at a certain (Ultrasound) The extinction cross section (or total cross sec- distance from the patient and the target within the patient is irra- tion) is defined as the time-averaged total absorbed and scattered diated with an external radiation beam. power divided by the time averaged incident intensity. The unit is Most external beam radiotherapy is carried out with photon in square meters, as implied by the name cross section. Physically beams, some with electron beams and a very small fraction with this cross section corresponds to the area of the incident wave that more exotic particles such as protons, heavy ions or neutrons. contains the amount of power that is absorbed and scattered by an Extracorporal elimination 351 Extrapolation ionisation chamber object. Consequently, it is the sum of the

scattering and absorp- copper with an opening at the side of the cathode. At the path of tion cross sections, defined analogously. the useful radiation the copper is replaced with Beryllium plate Related Articles: Scattering cross section, Absorption cross which can pass the x-rays with minimal absorption (Z = 4 for Be) section, Differential scattering cross section – see the figure in the article Stationary anode. Related Articles: Anode, Target, Stationary anode, Focal spot Extracorporal elimination effective (Nuclear Medicine) Extracorporal elimination is an approach used to minimise the radiation dose to normal tissue in patients Extraoral projection radiography undergoing radioisotope therapy. The ideal radionuclide used (Diagnostic Radiology) Extraoral projection radiography is a den- for therapy would accumulate selectively to the tumour cells tal x-ray investigation obtained with the image detector placed and spare normal tissue from most of the radiation dose. There outside the mouth. are a few suggested approaches on how to minimise the radia- Various types of extraoral radiograph can be acquired, both to tion dose to normal tissue. One of the methods suggested is the view the entire skull and to focus only the maxilla or the mandible. extracorporal elimination. A compound that binds with high 2D images can be acquired with a panoramic equipment affinity to the excessive targeting agents in systematic circula- (also known as an ortophantomograph) or with cephalometric tion can be introduced subsequent to targeting agent injection. equipment. The compound binds to the targeting agent and speeds up the Also, image detectors outside the mouth are always used in targeting agent clearance from the body, hence decreasing the dental CT or CBCT. These are 3D radiographic techniques, but radiation dose. One example of extracorporal elimination is the they should be mentioned because different 2D images can be E extraction of excessive biotinylated and radionuclide labelled reconstructed from the 3D. antibodies using an extracorporal column with immobilised Related Articles: Ortho pan tomography (OPG), CBCT avidin. Related Articles: Radionuclide uptake in tumour cells, Anti- Extrapolated range of electrons idiotype antibody technique (Radiotherapy) The extrapolated range of electrons (Rex) is found Further Reading: Carlsson, J., E. F. Aronsson, S-O. Hietala, by using a plot of the central axis percentage depth dose curve T. Stigbrand and J. Tennvall. 2003. Tumour therapy with radionu- (see Figure E.53). It is found by extrapolating the linear portion of clides: Assessment of progress and problems. Radiother. Oncol. the curve to the abscissa (x-axis, depth). The extrapolated range of 66 (1):107–117. electrons is sometimes also called the practical range, and while the two methods for calculating these are different the typical val- Extraction fraction ues will tend to be similar. Therefore the rules for approximating (Nuclear Medicine) Extraction fraction is the fraction of tracer in the practical range are a good approximation for the extrapolated blood which is extracted in one pass through the vascular bed of range, and so the practical range tends to be the more common a tissue. In terms of the tracer concentration in blood, it can be figure quoted. defined as the difference between arterial and venous blood con- Abbreviation: PDD = Percentage depth dose. centration divided by the concentration in arterial blood: Related Articles: Electron ranges, Electron maximum range, Electron therapeutic range, Electron practical range C E arterial blood - C = venous blood C arterial blood Extrapolation ionisation chamber (Radiation Protection) In an extrapolation ionisation chamber some of its parameters (e.g. distance between electrodes (spacing) The extraction fraction can also be thought of as the probability and thus volume) can be varied for measurement purposes. Then that a tracer molecule is retained in the tissue during a single pass. the measured values at different volumes/electrode spacing can Further Reading: Peters, A. M. 1998. Fundamentals of tracer be used by extrapolating the results to zero volume to determine kinetics for radiologists. Brit. J. Radiol. 71:1116–1129. the absorbed dose. Extrafocal radiation (Diagnostic Radiology) The extrafocal radiation is a phenomenon 100 in x-ray generation in the x-ray tube. Some of the accelerated ther- 90 mal electrons are scattered back from the anode (without creating 80 x-rays). These scattered electrons fall again on the anode (but out- side the actual focal spot) and create x-rays outside the effective 70 focal spot. This way the extrafocal radiation enlarges the effective 60 focal spot and leads to blurring the radiography – see the figure in 50 the article Focal spot effective. Some sources name this effect off- 40 focus radiation. In stationary x-ray tubes the scattered electrons 30 fall back on the anode stem (outside the target) and in this case the 20 extrafocal radiation is called also stem radiation. 10 Rex Various constructions of the x-ray tubes allow for minimising 0 this effect, which otherwise can reach to 15%–20% of the useful 0 10 20 30 40 50 radiation. Depth (mm) One possible construction minimising the extrafocal radiation is a metal ‘electron-capture’ hood, placed over the anode. This FIGURE E.53 An illustration of the definition of the extrapolated range hood retains the scattered electrons. The anode hood is made of of electrons. PDD (%) Extravascular 352 Extremity dosimeters The measurement of the absorbed dose Dm based on the Bragg–Gray principle: Dm = W ´ Sm ´ P[Gy] where W is equal to the average energy loss per ion pair in the gas Sm is a relative mass stopping power of the material to that of the gas P a number of ion pairs formed in the gas per unit mass The absorbed dose can be measured using a gas-filled ionisa- tion chamber with the cavity small in comparison with the range of primary or secondary charged particles of measured radiation. The cavity is filled with gas, e.g. air, and the wall of the chamber is made from the solid material air-equivalent i.e. Sm = 1. The measurement of the dose in biological tissue is performed using so called tissue E equivalent ion chambers. For the x-rays or gamma radiation of very low energies the measurement of the absorbed dose, e.g. in tissue, FIGURE E.54 Knee coil. is realised with use of the extrapolation chamber because of a very small ion current in a tissue equivalent ion chamber. The extrapola- tion chamber consists of a pair of electrodes for which spacing can need for RF excitation by the body coil and reduced SAR (specific be adjusted. The ion current per unit volume is measured for dif- absorption rate). ferent distances between electrodes to permit extrapolation to zero Related Article: RF coil chamber volume (zero spacing). The absorbed dose can be calcu- lated from the extrapolated ion current using Bragg–Gray principle. Extremity dosimeters Related Article: Ionisation chamber (Radiation Protection) Radiation workers may receive high doses Further Reading: Knoll, G. F. 2000. Radiation Detection to specific parts of the body because of the work that they are and Measurement, 3rd edn., John Wiley & Sons, Inc., New York, undertaking. Common sites for high doses to occupational work- p.144. ers are hands and eyes. In medicine, radiologists or cardiologists performing interventional procedures under fluoroscopic control Extravascular may receive hand and eye doses. Personnel dosimeters have been (Ultrasound) developed specifically for these sites. See Comet tail The main requirement for hand and eye monitoring is to pro- duce a dosimeter that is sufficiently small and flexible. The mate- Extremity coil rial which best suits this requirement is TLD. Personnel monitors (Magnetic Resonance) An extremity coil is an RF coil used in imag- for hands include rings and ‘finger stalls’, the latter used when it ing the knee, ankle, foot or wrist (Figure E.54). Most frequently the is the pulp of the finger that is most at risk of exposure, e.g. when term ‘extremity coil’ is used with reference to a knee coil, and ‘knee injecting radioactive substances. coil’ and ‘extremity coil’ are often used interchangeably. Some Tapes containing TLD are also available to be attached to extremity coils are designed either for knee or for foot/ankle imag- glasses or directly to the forehead or neck when monitoring for ing, whereas other designs combine both capabilities. eyes or thyroids is required. As volume coils, knee coils can be designed in a transmit/ Related Articles: Finger ring dosimeter, Integrating dosim- receive (T/R) configuration. Use of a T/R coil design avoids the eter, Thermoluminescent dosimeter (TLD) F FAIR 2. The diagnosis indicates a positive answer (disease) to a (Magnetic Resonance) See Flow-sensitive alternating inversion patient that does not have a disease. recovery 3. The diagnosis indicates a negative answer (no disease) to a patient that has a disease. Falling-load generator 4. The diagnosis indicates a negative answer (no disease) (Diagnostic Radiology) This type of x-ray generator is specially to a patient that does not have a disease. designed to produce short x-ray exposures (useful for radiograph- ing moving organs). It does not use the normal way of producing the The preceding situations are often labelled as follows: exposure with constant anode current (mA) but changes the anode • 1 is the true positive (TP) current during the exposure (Figure F.1). This way the generator • 2 is the false positive (FP) uses the maximal possible mA (following the maximal permissible • 3 is the false negative (FN) mA from the tube loading chart). Applying the maximal mA allows • 4 is the true negative (TN) the same mAs to be achieved for the minimal exposure time. Figure F.1 gives an example presented at a set of typical tube The fraction of true positive to all cases is called the sensitivity, load chart. TP/(TP + FN), and the fraction true negative to all cases is called If an exposure with 120 kV and 10 mAs is required, it can only specificity TN/(TN + FP). A ROC analysis determines the frac- F be achieved by applying 80 mA for 125 ms (80 mA × 0.125 s = 10 tions of 1–4. The result is usually plotted as a ROC curve. The fol- mAs). If higher mAs are used, the necessary time will go above lowing figure shows such curves of two imaging modalities where the maximal permissible load curve for 120 kV (100 mA and 100 modality B is better than modality A. ms cross above the line). If falling mode operation is used, initially the exposure begins 1.0 with 200 mA for 10 ms (equal to 2 mAs), continues with 150 mA for 20 ms (equal to 3 mAs), and then continues with 100 mA for 50 ms (equal to 5 mA). This way 2 + 3 + 5 = 10 mAs are delivered Modality B for 80 ms – i.e. 64% of the time of the first exposure during the operation, the x-ray tube load is always kept below the maximum permissible load. Modality A Despite the shorter exposure time, these generators could lead to focal spot enlargement (hence decreased resolution). This is due to the quick heating of the anode during this high mA opera- tion. Falling load generators are more expensive and usually work in radiographic systems with automatic exposure control. The production of more powerful x-ray tube gradually decreases the need of these x-ray generators. Related Articles: High voltage generator, Tube rate chart 0.0 Falling load with raising kV generator 0.0 1.0 False positive fraction (Diagnostic Radiology) Falling load with raising kV is a histori- cal mode of operation of the X-ray generator, where during the exposure, a simultaneous decrease of mA and increase of kV Related Articles: True positive, False positive, True negative take place. This leads to further shortening the exposure time and improvement of radiography contrast with less absorbed dose. False positive (FP), ROC The method is not used with contemporary medium frequency (General) See False negative generators. Related Article: Falling load generator FAMPO (General) The Federation of African Medical Physics False negative (FN) Organizations (FAMPO) was founded in 2009 as a regional (General) The diagnosis of a disease using some kind of imaging organisation of IOMP. As of 2019 the Federation consists of nine modality, such as a scintillation camera, involves some kind of national member organisations, representing about 700 physicists uncertainty in the decision of whether a patient has an abnormal- and engineers working in the field of medical physics. ity or is normal. There can be four alternatives: Since its inauguration the main objective of FAMPO has

been 1. The diagnosis indicates a positive answer (disease) to a to harmonise and promote the best practice of medical physics in patient that has a disease. the African Region. 353 True positive fraction Fan beam 354 Fano’s theorem 800 For image reconstruction, the fan beam data are re-binned, so 60 kVp as to obtain parallel projection data sets. 80* kVp 700 80 kVp Related Article: Computed tomography 100 kVp 120 kVp Fan beam collimator 600 (Nuclear Medicine) A converging collimator where the holes con- verge to a focal line. Commonly used in brain imaging. 500 A fan beam collimator has septa arranged such that each row of holes in the radial direction has its own focal point (Figure F.3a). Sequential rows are stacked and evenly spaced such that they are 400 parallel to one another along the z-axis (Figure F.3b). As such, the fan beam collimator has characteristics of a converging col- 300 limator in the radial direction, and a parallel hole collimator in the z-direction. This is in contrast to a cone beam collimator design, which features converging holes in both the radial and 200 z-directions. Each row of holes in a fan beam collimator therefore provides its own independent projection profile. As the fan beam collimator behaves like a converging collima- 100 tor in the radial direction, image magnification is present in this direction (see converging collimator). The FOV also decreases 0 with distance from the collimator. 10 100 1000 This configuration allows data from smaller structures such F Exp. time (ms) as the heart and brain to use the maximum surface area of the crystal, hence exhibiting better resolution and sensitivity for small FIGURE F.1 Mode of operation of a falling load generator, presented FOV imaging applications when compared with parallel hole over a tube rate chart (see the eponymous article). collimator designs. Conversely, cone beam collimators magnify the image in both the radial and z-directions. FAMPO includes the medical physics societies of the follow- SPECT data acquired with a fan beam collimator, cannot ing countries: Algeria, Egypt, Ghana, Morocco, Nigeria, Niger, be directly reconstructed using standard methods designed for South Africa, Sudan, Tunisia. parallel hole data reconstruction. It is either necessary to sort the Further Reading: J. Med. Phys. Int. Dec 2019 (with focus on fan beam data into parallel beam data before reconstruction, or FAMPO) – www .mpijournal .org to reformulate the FBP reconstruction algorithm to incorporate Hyperlinks: www .fampo -africa .org fan beam data. Related Articles: Converging collimator, Parallel hole colli- Fan beam mator, Cone beam collimator (Diagnostic Radiology) The first clinical CT scanners employed Further Readings: Cherry, S. R., J. A. Sorenson and M. E. pencil, parallel x-ray beam geometry to acquire data. Modern CT Phelps. 1986. Physics in Nuclear Medicine, 4th edn.; Design and scanners employ a fan beam of x-rays in the scan (x–y) plane. The Clinical Utility of a Fan Beam Collimator for SPECT Imaging angle of the fan beam, subtended by the detector arc, is about 40°– of the Head; Tsui, B. M. W., G. T. Gullberg, E. R. Edgerton, 50°, so that a whole object up to approximately 50 cm in width can D. R. Gilland, J. Randolph Perry and W. H. McCartney. 1986. be encompassed by the beam at the isocentre (Figure F.2). This J. NucI. Med. 27:810–819; Van Audenhaege, K., R. Van Holen, removes the need for translation between angular projections (see S. Vandenberghe, C. Vanhove, S. D. Metzler and S. C. Moore. the article Computed tomography) and reduces the time for data 2015. Review of SPECT collimator selection, optimization, acquisition. and fabrication for clinical and preclinical imaging. American Association of Physicists in Medicine. Fano’s theorem Fan angle (Radiotherapy) The Bragg–Gray cavity theory deals with the determination of the absorbed dose at a specified point in a medium from a measurement with a radiation detector. The cav- ity theory considers a medium uniformly irradiated by photons and of sufficient dimension that charged particle equilibrium Fan beam (CPE) is established at the place where the cavity is introduced. The CPE exists for a volume v if each charged particle of a given type and energy leaving the volume is replaced by an identi- cal particle of the same energy. Some conditions must exist to realise the CPE condition including that the density of the irradi- ated medium is homogeneous. In order to measure the absorbed dose in a medium, it is necessary to introduce a radiation detector into it and, in general, the device will not be of the same material FIGURE F.2 Diagram of x-ray fan beam on a CT scanner. (Courtesy of and density as the medium. This will be the case, for example ImPACT, UK, www .impactscan .org) of an ionisation chamber introduced in a medium to perform Ia (mA) Far zone 355 Faraday cup F FIGURE F.3 (a) Fan beam collimator in the radial direction. (b) Fan beam collimator in the z-direction. measurements. Equilibrium may also be established in media external sources of radio frequency (RF) radiation would inter- which are not completely homogeneous but which exhibit varia- fere with the sensitive measurement of the weak nuclear magnetic tion in density. This is affirmed by Fano’s theorem (see Further resonances of the spins. The cage is constructed by lining the Reading) which states that: in a medium of given composition walls of the room with a continuous conductive surface made of exposed to a uniform flux of primary radiation (such as x-rays sheets of copper, aluminium or steel. The conductive metal cage is or neutrons), the flux of secondary radiation is also uniform and not normally visible as the room is lined with a decorative finish. independent of the density of the medium as well as of the density The cage continues through the observation window of the room variations from point to point. Fano’s theorem requires that the as a conductive mesh set into the window glass. The door frame interaction cross sections should be independent of density. This must be lined with conductive contact strips to ensure continu- requirement is never fulfilled completely because of the polarisa- ation of the conductive cage through the room door. Power and tion effect due to the polarisability of the medium which causes signal cables are routed to the MRI via a penetration panel which the mass stopping power of charged particles to be lower in a con- uses dedicated filters to prevent penetration of RF signal into the densed medium than in a gas of identical atomic composition. room. Any external tubing or fibres must be transferred into the Further Reading: Fano, U. 1954. Note on Bragg–Gray cav- room through conductive tubes called waveguides. The diameter ity principle for measuring energy dissipation. Radiat. Res. and length of the waveguides are chosen to prevent transmission 1:237–240. of RF signal with frequencies in a range that could interfere with imaging (Figure F.4). Far zone (Ultrasound) See Diffraction Faraday cup (Radiotherapy) A Faraday cup is a metal cup used to detect Farad charged particles in a vacuum. (General) The units of capacitance in the International System of Charged particles hitting the inner surface of the cup cause a Units (SI) are C V−1 or farads (F). If a capacitor carries charge Q current to flow between the cup and ground. The current can be coulombs on either plate when the potential difference between measured and used to determine the number of ions or electrons its terminals is V volts, then its capacitance is Q/V farads. that entered. Abbreviation: SI = International System of Units. As the associated currents are very small, shielding and pre- Related Article: Coulomb amplification electronics are required. Ideally the cup should measure the total intensity of the ion Faraday cage beam. However, ions entering typically have energies signifi- (Magnetic Resonance) A Faraday cage is a conductive enclosure cantly higher than the energy needed to remove an electron from designed to prevent the penetration of external electromagnetic the cup material (e.g. copper, stainless steel, graphite, carbon) to radiation to a sensitive device housed within it. In MRI, the room the vacuum. If a secondary electron leaves the cup, this varies housing the scanner is built as a Faraday cage. Without the cage, the charge on the cup: it appears that an additional positive ion Faraday shield 356 Farmer chamber (a) (b) F FIGURE F.4 (a) Conductive contact strip at door opening of MRI scan room. (b) Two waveguides in the wall of an MRI scan room. has entered. This represents a source of error. Another source of conducting mesh that is moulded inside the glass. The door has error arises when a charged particle entering the cup is scattered flexible metallic filaments all around its edges that connect it to back at 180 degrees. The particle may enter the cup again but is the rest of the cage when it is closed. indistinguishable from a fresh incident particle. If any part of the shield is damaged, for example by running a The number of secondary electrons depends on ion mass, ion nail through it, or if the door is left open, the attenuation no longer energy, cup material, angle of incidence and ion nature (mona- works and image artefacts may occur due to radio signals that the tomic/polyatomic). The Faraday cup is a detector with limited MR-scanner picks up from the outside environment. sensitivity and dynamic range. Related Articles: Magnetic resonance imaging (MRI), Radiofrequency Faraday shield (Magnetic Resonance) The term Faraday shield refers, in Farmer chamber the context of MRI, to the conducting enclosure of the entire (Radiotherapy) The Farmer chamber is used to determine radia- examination room where the scanner is installed. It may also tion dose by measuring the charge accumulated between two be denoted radio frequency (RF) shield since its purpose is to electrodes when the chamber is irradiated. One electrode is the prevent electromagnetic radiation at the radio frequency range cylindrical graphite outer electrode and the other is a central alu- and particularly at the resonance frequency of the MR-scanner minium electrode, see Figure F.5a. The chamber dimensions are from passing in or out. Since the field strengths of clinical shown in Table F.1. MR-scanners may typically be 0.2–3.0 T, the frequency range Farmer chambers may be supplied with a build-up cap that needs to be heavily attenuated by the shield is (according to (Figure F.5b). The chamber is placed inside this to achieve CPE the Larmor equation) on the order of 10–130 MHz. The attenu- when irradiated in air in a cobalt 60 beam; however, the build-up ation in this frequency range for a well-functioning shield can cap is not used when measuring doses routinely in the clinic. be 90 dB or better. This device can be used for relative or absolute dose measure- The shield is normally constructed using copper foil mounted ments, and because the chamber is open to the atmosphere, a tem- on wooden sections. Stainless steel may also be used, but is more perature and pressure correction has to be made to readings where difficult to work with. The sections are put together at the inside absolute measurements are required. of the walls, floor and roof of the scanner room and every part needs to be electrically connected to one solid cage covering the room. The entire cage also needs to be electrically insulated from the rest of the building. Special solutions such as frequency fil- TABLE F.1 ters are required at spots where the shield needs to be penetrated. Chamber Dimensions Therefore, there is a penetration panel where for example cables Nominal cavity volume 0.6 cm2 needed for power and communication with the scanner are pass- ing. There are also waveguides, normally brass tubes with a spe- Cavity length 24 mm cific length/diameter relationship, through which non-conductive Cavity diameter 6.25 mm cables such as fibre optics may enter the scanner room. Windows Graphite wall thickness 0.5 mm need to be shielded too, which is accomplished by the use of a Fast CT 357 Fast imaging with steady state precession (FISP) Graphite outer electrode sweeping across different targets. This method is implemented in electron beam tomography in medical CT. The third approach is Graphite outer electrode Aluminium inner electrode the switched source technique in which static sources are com- bined with multiple

detectors and are switched electronically. Fast field echo (FFE) (Magnetic Resonance) FFE refers to a gradient echo generated at an echo time after the radiofrequency excitation pulse by apply- (a) Dural ing a bipolar gradient pulse in the readout (frequency) direction. Since the echo is formed during the free induction decay (FID), the echo signal exhibits a T2* weighting. The term FFE can be used interchangeably with the term gradient echo. See related article which also includes a short description of the basic gradi- ent echo pulse sequence. Related Articles: Fast low angle shot (FLASH), Free induc- tion decay (FID), Gradient echo Fast Fourier transform (General) The FFT is a powerful efficient algorithm for comput- ing the discrete Fourier transform (DFT). For N data points, it reduces the number of computations from the order of N2 to the (b) order of N log2(N). If the data are 1024 long, the saving is a factor F of 100. An FFT produces exactly the same result as evaluating the FIGURE F.5 Farmer ionisation chamber. (a) Scheme of electrodes. (b) DFT directly. The Farmer chamber and its build up cap. Most FFT algorithms operate in optimal efficiency when N can be factorised easily, for example if N is a power of 2. The Cooley–Tukey FFT algorithm is such an example. Abbreviations: FFT = Fast Fourier transform and DFT = Further Reading: Mayles, P., A. Nahum and J.-C. Rosenwald. Discrete Fourier transform. eds. 2007. Handbook of Radiotherapy Physics – Theory and Related Articles: Fast Fourier transform, Discrete Fourier Practice, Taylor & Francis, New York, pp. 290–292. transform Further Reading: Bracewell, R. 1986. The Fourier Transform Fast CT and Its Applications. McGraw-Hill Book Co., Singapore. (Diagnostic Radiology) Across both the medical and non-destruc- tive material inspection industries there has been always a need for faster tomographic imaging techniques. This need stems both Fast imaging with steady state precession (FISP) from an increase in demand and requirements to study dynamic (Magnetic Resonance) The free induction steady precession processes such as cardiac scanning in medical imaging and fluid (FISP) belongs to the family of steady state free precession flow in pipes in the petroleum industry. (SSFP) sequences. This is a gradient echo sequence that combines For conventional medical scanners, increasing the rotation short repetition times and low flip angles to establish a steady speed of the source and detector gantry is the most obvious solu- state of the transverse magnetisation. There are several variants tion. This, however, will always be limited by the mechanical in the family of steady state free precession depending upon how issues related to rotating a very heavy gantry (up to 6000kg). the gradients are designed to be balanced in one, two or all three Dual-source medical CT scanners have been developed with an gradient directions. in-plane temporal resolution of up to 75 ms and just over three Figure F.6a shows the timings of a FISP sequence, i.e. a SSFP rotations per second. sequence balanced in the phase encode direction but unbalanced For laboratory micro-CT systems where the sample is rotated along the readout-direction. The difference between a FISP instead, scan speeds are also limited by the sample stage rotation sequence and a FLASH sequence is the balanced gradients in the speeds and the lower dose rate of the small focal spot size source. phase encode direction and the lack of spoiler gradients or spoil- For these systems, higher dose rate sources have been developed ing RF pulses. For short TRs and large flip angles, the SSFP-FID based on liquid metal jet technology leading to a potential order sequence provides images of a mixed T1/T2 contrast, using short of magnitude improvement in scan speed. In this type of source TRs, low flip angles and short TEs; the contrast is proton den- the conventional anode is replaced by a liquid-metal jet where sity (PD) weighted, but with a long TE, the sequence gives T2*- the anode is continuously regenerated and already in the molten weighted images. stage. This removes the classical power limit restriction for an Figure F.6b shows a true FISP sequence, in which all three x-ray source where the anode is damaged by the electron beam. gradient directions are balanced. The image contrast of such a For industrial CT systems developments have been focused in balanced SSFP sequence is related to the T2/T1 ratio. Therefore, three main areas, all with no moving parts. The first is the use of tissues with a high ratio, such as bile, blood and fat appear bright. multiple static sources simultaneously each opposite a block of The sequence allows ultra-fast imaging in all areas of the body detector elements. The second approach is to implement an elec- and is regularly used in cardiac MRI, MRA, foetal imaging in tronically moving source via the use of one or more electron guns utero and for MR-guided interventional procedures. Fast kV switching 358 Fast timing techniques for single-channel analysers RF RF Gs Gs Gp Gp Gf Gf Echo time (TE) Echo time (TE) Repetition time (TR) Repetition time (TR) (a) (b) FIGURE F.6 Schematic illustration of (a) a partially refocused SSFP-FID sequence (FISP) and (b) a fully refocused SSFP-FID (true-FISP) sequence. Acronyms for the SSFP-FID sequences are as follows: pulse, in order not to interfere with the transverse magnetisa- Balanced in one direction: GRE /GRASS (General Electric), tion produced by the next excitation pulse (see related article F FFE (Philips) and FISP (Siemens). Gradient echo which also includes a short description of the Balanced in three directions: FIESTA (General Electric), bal- pulse sequence). anced FFE (Philips) and true-FISP (Siemens). Related Articles: Fast field echo, Flip angle, Gradient echo Related Articles: Fast low angle shot (FLASH), Flip angle, Steady state free precession (SSFP) Fast spin echo (FSE) Further Readings: Nitz, W. 2002. Fast and ultrafast non- (Magnetic Resonance) FSE is an MRI pulse sequence character- echoplanar MR imaging techniques. Eur. Radiol. 12:2866–2882; ised by the encoding of multiple k-space lines in a train of spin Scheffler, K. 1999. A pictorial description of steady-states in rapid echoes (Figure F.8). For example, if three echoes are sampled magnetic resonance imaging. Concepts Magn. Reson. 11(5):291– after each 90° pulse, the total acquisition time will be decreased 304; Scheffler, K. and J. Hennig. 2003. Is trueFISP a gradient- by a so-called ‘turbo factor’ of three compared to a single-echo echo or a spin-echo sequence? Magn. Reson. Med. 49:395–397. spin echo sequence with identical parameters. The number of echoes is also known as the echo train length, ETL. Fast kV switching The time period between the excitation pulse and the acquisi- (Diagnostic Radiology) Fast kV switching is a method in which tion of the central k-space line is denoted the effective echo time, the tube voltage is modulated between high kV (140 kV) and low TEeff, and this parameter can be changed by adjusting the echo kV (80 kV) so that the projection from every angle is measured spacing (the time between echoes) and by ordering of the k-space two times with different x-ray spectra. It is one of the methods for line sampling strategy. Since the echoes associated with the cen- performing dual energy CT. tre of k-space determine the contrast in the images, low TEeff The GE Healthcare marketing name for fast kV switching is gives T1 – or PD weighting (depending upon the choice of repeti- ‘gemstone spectral imaging’ (GSI), where gemstone is in refer- tion time) while high TEeff (late encoding of the echoes in the ence to the garnet crystal scintillator material in the detector ele- centre of k-space) in combination with a long repetition time will ments. Gemstone has a shorter decay time of the scintillation light lead to T2-weighted images. output compared to the standard gadolinium oxysulfide scintil- Since a high number of powerful 180° radiofrequency pulses lators that are generally used in CT scanners. With a fast decay are applied, FSE sequences tend to have a high specific absorption of the light signal, the sampling rate of projection rays can be ratio (SAR). SAR limits may be met by reducing this flip angle increased, thereby achieving both high temporal and spatial reso- and using hyper-echo techniques. lution and it is helpful for conducting a fast kV switching scan. Other names for variants of this sequence are turbo spin echo (Figure F.7). (TSE), rapid acquisition relaxation enhancement (RARE) and Related Articles: Dual energy CT half-Fourier single-shot turbo spin echo (HASTE). Related Articles: Echo train length, Half acquisition single- Fast low angle shot (FLASH) shot turbo spin echo (HASTE), Rapid acquisition relaxation (Magnetic Resonance) FLASH commonly denotes a spoiled enhancement (RARE), Turbo spin echo (TSE) gradient echo pulse sequence. In this sequence, the amount of Further Reading: Bernstein, M. A., K. F. King and Z. J. Zhou. transverse magnetisation is conserved by establishing a steady 2004. Handbook of MRI Pulse Sequences, Elsevier Academic state. For this, the flip angle of the radiofrequency excitation Press, Burlington, MA. pulse is smaller (typically 10°–70°) than for a spin echo pulse sequence (typically 90°). This makes it possible to use much Fast timing techniques for single-channel analysers shorter repetition times and correspondingly shorter imag- (Radiation Protection) A timing single-channel analyzer marks ing times. The remaining transverse magnetisation after each the arrival time of detected pulses called an event time. The time echo is spoiled, for example by applying a random spoiling resolution depends on the detector (the best is for the fastest detec- gradient or by randomly varying the phase of the excitation tors, e.g. scintillation detector) and on the ratio of maximum to Fast timing techniques for single-channel analysers 359 Fast timing techniques for single-channel analysers F FIGURE F.7 Illustration of fast kV switching method. RF GS GP GF Effective echo time (TEeff) Echo train length (ETL) FIGURE F.8 Illustration of a FSE pulse sequence. RF denotes the radio frequency pulse and GS, GP and GF are the slice-selective, phase and fre- quency encoding gradients, respectively. minimum pulse height, i.e. dynamic range of pulses. However, it (ARC) timing (e.g. for germanium detectors where the rise time is limited by the following factors: variations are large), extrapolated leading edge timing (ELET) a. Jitter for constant amplitude of input pulses caused by and first photoelectron timing (FPET) for scintillation detectors. pulse statistical fluctuations and input noise The choice of the method depends on the pulse amplitude range b. Amplitude walk effect time slewing for variable ampli- (narrow or wide) and its shape as well as its rise time. The best tudes of input pulses timing is provided for detectors having fast and non-variable rise Both the factors are shown in Figure F.9 in leading edge trigger- time. ing. The leading edge (LE) triggering is the easiest direct time The timing single channel analysers (SCAs) are used in coin- pick-off method using a fixed discrimination level (threshold) to cidence systems (e.g. events detected by two PM tubes occurred mark out the time at which a pulse crosses it. simultaneously) or in time spectroscopy (time interval measure- Besides the leading edge triggering, there are other time pick- ment) in the range of 10−6–10−12 s. The time interval Δt mea- off methods such as zero crossover and fast crossover timing, con- surement may be performed with the time-to-amplitude converter stant fraction (CF) timing, amplitude and rise time compensated (TAC). This device produces an output pulse with an amplitude Fat 360 Fat suppression Time jitter Amplitude walk SCA window Time (a) (b) FIGURE F.9 Jitter (a) and amplitude walk (b) in leading edge (LE) timing method. Detector 1 Preamplifier Linear Timing amp SCA Time Det. bias pick-off F Coincidence Solar Det. bias Detector 2 Preamplifier Linear Timing Variable amp SCA delay Time pick-off FIGURE F.10 Block scheme of a two-coincidence measuring system for pulse amplitude selection and timing. proportional to Δt between input start and stop pulses. The ampli- Fat saturation (FATSAT) tude distribution may be measured by a SCA or multi-channel (Magnetic Resonance) Fat saturation, commonly referred to analyzer (MCA). The example of a coincidence measuring system as FATSAT, selectively saturates lipid (fat) prior to an excita- and time spectroscopy is presented in Figure F.10. tion RF-pulse. This technique employs a frequency-selective Further Reading: Knoll, G. F. 2000. Radiation Detection and RF-prepulse and a subsequent spoiler gradient that saturates the Measurement, John Wiley & Sons, Inc.,

New York. fat resonances while leaving those of the water unaffected. This results in a negligible signal from the lipid protons in the images. Fat Since this technique relies on frequency-selective RF-prepulses (General) The term fat refers to lipids that are solid at room tem- prior to the imaging/spectroscopy sequence, the fat saturation perature. Fats are hydrophobic triglyceride molecules composed technique can be used with almost any pulse sequence. of one glycerol and three fatty acid molecules. Fats can either be Effects of magnetic field inhomogeneities can, however, result saturated, where the fatty acid contains no double bonds, or unsat- in a poor suppression due to the shift in the lipid resonance fre- urated, where the fatty acid molecule does contain double bonds. quency. For example, fat saturation might be less suitable in In the human body, all excess carbohydrates, proteins, fats and inhomogeneous volumes of tissue or in regions with metallic oils are stored in adipose tissue. Fat has twice the energy storage prosthesis that causes an alteration in the local magnetic field and capacity per kilogram compared to carbohydrate. can result in unwanted suppression of off-resonance water signal. Properties of Fat Based Tissues for Medical Imaging: Fat- Related Articles: CHEMSAT (chemical selective saturation), based tissues are visible on both x-ray and magnetic resonance Fat suppression images. They have short T1s and T2s of between 100–150 ms and 10–100 ms, respectively. The CT number of fat-based tissues is Fat suppression about 120 Hounsfield units. (Magnetic Resonance) Fat suppression can be achieved by three Related Articles: Computed tomography, Fat nulling, Fat methods: fat saturation, inversion recovery imaging and opposed saturation, Fat suppression, Hounsfield scale, Magnetic resonance phase imaging. Each of the three methods has advantages and imaging, T1, T2, Tissue, X-ray disadvantages, which will be described in the following. Fat saturation is achieved by applying a frequency selective Fat nulling RF-pulse and a subsequent spoiling gradient, which suppresses (Magnetic Resonance) See Fat suppression the fat resonances while leaving the water signal unaffected. Fat FATSAT (fat saturation) 361 Ferromagnetic materials saturation can be used with almost any pulse sequence since it is Feedback performed as a magnetisation preparation experiment before the (General) Feedback describes the arrangement where part of the excitation RF-pulse of the imaging sequence. Since fat saturation output is fed back to the input to achieve automatic self-regulation relies on a frequency-selective saturation pulse, effects of mag- of an electrical, mechanical or biological system. In automa- netic field inhomogeneities will result in a poor suppression due to tion, the use of feedback is fundamental control mechanism for the shift in lipid resonance frequency and an unwanted suppres- machinery. Feedbacks are also widely used in electronic control sion of water resonances. systems. Short TI inversion recovery (STIR) is based on the application In negative feedback systems, if the output signal increases, of an inversion RF-pulse before the imaging sequence. Since the the feedback circuit reduces the signal in the input (and vice T1 relaxation time of fat is shorter than for water, the longitudi- versa), thus stabilising the system. For example, in voltage and nal magnetisation of fat will recover faster than that of water. current regulators, part of the output is used as a control input, The excitation RF-pulse of the imaging sequence is applied at providing self-regulation. an inversion time (TI) when the longitudinal magnetisation of In positive feedback systems, if the output signal increases, fat crosses zero ensuring suppression of fat signal. Usually, this the feedback circuit increases the signal in the input, this way occurs at 0.69 times the T1 of fat. For example, the optimal for creating a generator of oscillations. Usually such systems have fat suppression at 1.5 T is TI = 130–170 ms. The STIR pulse internal sensor for overloading, which interrupts the oscillations. sequence is the only method that is insensitive to inhomogene- Feedback control systems of great complexity also exist in liv- ities and can be used with low-field-strength magnets. However, ing organisms. the method is non-specific since water with T1 similar to that of fat also will be suppressed. The inversion recovery preparation Ferromagnetic materials introduces an additional T1 contrast, which can be suboptimal if (Magnetic Resonance) Like the mass and the electrical charge a PD or T2-weighted contrast is required. In addition, the signal- of a particular element or material, magnetism is a fundamental to-noise ratio is also reduced since the longitudinal magnetisa- property of matter. All materials interact in some form with an tion of the water will not have fully recovered at the time of the F external magnetic field. The different types of magnetic materials excitation. are usually classified in terms of their susceptibility or permeabil- The opposed phase technique is based on the different reso- ity. The permeability is defined as μ = B–/H–, and the susceptibil- nance frequencies between water and fat signal, which results in ity is defined as χ = M–/H–, where B is the magnetic induction, H phase differences at different echo-times in gradient echo imag- the magnetic field and M the magnetisation, that is the magnetic ing. Usually, images at two different echo-times are acquired in dipole density. The apparent magnetisation of an atom is given by which water and fat signals will interfere destructively (opposed phase) and constructively (in phase). At 1.5 T, the opposed phase occurs at TE = 2.26 ms, whereas in phase at TE = 4.52 ms. The M = cH, thus B = m0(H + M) = m0 (1+ c) H in-phase image (Iin) is the sum of the fat and water signal, i.e. Iin = Iwater + Ifat, whereas the opposed-phase image (Iopp) is the where μ0 = 4π × 10−7 H m−1 is the permeability of free space (in SI difference image, i.e. Iopp = Iwater − Ifat. A fully fat-suppressed units), which is a universal constant. image can be generated from two images with different TE as The magnetic susceptibility of a material as detected by NMR Iwater = (Iip + Iopp)/2. If only one image is acquired, Iopp can is related to the ability of an external magnetic field to affect be used as a fat-suppressed image assuming the same amount of nuclei of a particular atom. This is strongly related to the electron water and fat signal in each voxel. The technique is simple and configurations of that atom. For example, the nucleus of an atom fast, but the one-image method also affects the water signal, while which is surrounded by paired electrons is less affected by an the difference method with two images is sensitive to field inho- external magnetic field than a nucleus with unpaired electrons. mogeneities. Recently, techniques were proposed that can com- Materials having a negative value of susceptibility are called dia- pensate for field inhomogeneities by acquiring three images at magnetic, when the susceptibility has a value between 0 and 1 different echo times. materials are called paramagnetic and ferromagnetic if the sus- A fourth alternative to avoid the fat signal is to only excite the ceptibility is greater than one. water using a frequency selective excitation pulse, but this is not a Paramagnetism: Unpaired electrons in paramagnetic mate- fat suppression method per se. rials induce small magnetic moments. With no external mag- Related Articles: Fat saturation, Inversion recovery, Short tau netic field, these magnetic moments have a random orientation, inversion recovery (STIR) and, therefore, they cancel each other out. In the presence of an Further Readings: Delfaut, E. M., J. Beltran, G. Johnson, J. external magnetic field, paramagnetic materials align with the Rousseau, X. Marchandise and A. Cotten. 1999. Fat suppression direction of the field and so the magnetic moments add together, in MR imaging: Techniques and pitfalls. RadioGraphics 19:373– resulting in a local increase of the magnetic field. 382; Glover, G. H. and E. Schneider. 1991. Three-point Dixon Diamagnetism: Diamagnetic materials present a zero net technique for true water/fat decomposition with B0 inhomogene- magnetic moment with no external magnetic field. Diamagnetic ity correction. Magn. Reson. Med. 18(2):371–383. materials present no elementary magnetic dipoles. This is because the electron currents caused by their motions add to zero. When FATSAT (fat saturation) an external magnetic field is applied, small magnetic moments (Magnetic Resonance) See Fat saturation (FATSAT) are induced, which counteract the applied field (Lenz’ law). This determines a subsequent slight decrease in magnetic field within Feature extraction the sample. Diamagnetic materials have negative magnetic sus- (General) Feature extraction refers to the process whereby a set ceptibilities and are therefore slightly repelled by the magnetic of measured features within an object is compared to established field. Diamagnetic effects appear in all materials. However in a criteria in order to classify the object. material which possesses both diamagnetic and paramagnetic Ferromagnetism 362 F-factor properties, the positive paramagnetic effect exceeds the negative with unpaired spins; these tiny magnetic dipoles are aligned paral- diamagnetic effect so the material appears paramagnetic. lel to each other within small regions of the material to form areas Ferromagnetism: If a ferromagnetic material is in the pres- of stronger magnetism. Compared to a paramagnetic material, a ence of an external magnetic field, the results are strong align- permanent magnetisation exists in the ferromagnetic material ment and attraction. The material retains its magnetisation even after the externally applied field is removed. Examples of ferro- when the external magnetic field has been removed, and, there- magnetic materials are iron (26), cobalt (27) and nickel (28). The fore, they are permanently magnetised and subsequently become numbers in the parenthesis are the atomic number of the material. permanent magnets. The magnetic field in permanent magnets If a ferromagnetic object (e.g. an implant) is placed in an MRI can be hundreds or even thousands of times greater than the scanner, the object will distort the homogeneity of the main mag- applied external magnetic field. netic field and cause susceptibility artefacts in the images. In Figure F.11, a qualitative representation of lines of forces of Related Articles: Susceptibility, Paramagnetism the magnetic induction is shown for the paramagnetic, diamag- netic and ferromagnetic substances. In ferromagnetic substances, Ferrous sulphate dosimetry the atomic dipoles tend to align in the same direction over regions (Radiotherapy) See Fricke dosimeter or domains containing millions of atoms forming a small mag- netic dipole. F-factor The interactions between a ferromagnetic object and an exter- (Diagnostic Radiology) The F-factor is a conversion factor in radi- nal magnetic field may have two effects. If the net dipole forms an ology between the exposure defined as the amount of ionisation in angle to the main magnetic field, it tends to align to the main field air and the absorbed dose in tissue caused by the x-ray exposure. direction under the action of a torque. The object may also experi- The relationship between ionisation in air and in tissues is not ence a force along the gradient of the magnetic field. Both torque fixed but depends on the type and energy of the radiation and on and direct attraction can affect any ferromagnetic object inside the type of tissue involved: F the patient as well as outside in presence of an external magnetic field. A metallic cylinder, 3 mm in diameter and 75 mm long with F-factor = Absorbed dose/Exposure a mass of 4.2 g, which forms an angle of 45° to a field strength of 1 T, requires a force of around 6000 N (600 g) applied at each end to prevent a twist. If the cylinder is free to move towards the In the United States, the traditional units used are magnet, it will achieve a final velocity of 17 m s−1. The velocity is • Ionisation in air: Roentgen (R) (1 R = 2.58 × 10−4 C kg−1) irrespective of the shape of the object and depends only on object • Absorbed dose: rad (1 rad = 0.01 Gy) mass. In these units, the F-factor for diagnostic x-rays varies between Ferromagnetism approximately 1 for soft tissues and up to 4 for bone: (Magnetic Resonance) The spin of the electron results in a mag- netic dipole moment that creates a magnetic field. In atoms with • Most other countries use the MKS units: unpaired spins, a net magnetic moment exists even in the absence • Ionisation in air: Coulombs per kilogram (C kg−1) of an external field.

Ferromagnetic materials contain many atoms • Absorbed dose: Joules/kilogram (Gy) Qualitative representation of lines of forces Paramagnetic substance Substance outside the magnetic Uniform magnetic field field Diamagnetic substance Ferromagnetic substance Magnetic pole FIGURE F.11 Qualitative representation of lines of forces of the magnetic induction is shown for the paramagnetic, diamagnetic and ferromagnetic substances. FFAG (Fixed field alternating gradient) accelerator 363 F ick’s method In this case, the MKS F-factor varies between approximately 40 clip, avoiding the use of a conductive cable near the MRI bore. (soft tissues) and 160 (bone). Data collection from patient monitoring systems placed in MRI Further Reading: Bushberg et al. 2002. The Essential scan rooms to remote display consoles is typically via fibre optics, Physics of Medical Imaging, Lippincott Williams & Wilkins, as is control of contrast injectors. Philadelphia, PA. Hyperlink: http: / /www .osha .gov/ SLTC/ radia tioni onizi ng /in Fibre optic taper troto ioniz ing /i onizi nghan dout. html (Diagnostic Radiology) One of the methods of indirect digital acquisition of x-rays is with charge-coupled device (CCD) FFAG (Fixed field alternating gradient) accelerator detectors. CCD-based detectors consist of three components: (Radiotherapy) See Fixed field alternating gradient (FFAG) accelerator. 1. A phosphor layer that converts incident x-rays to light photons FFE (fast field echo) 2. A light guide with collects and directs these photons (Magnetic Resonance) See Fast field echo (FFE) 3. Multiple CCD arrays to covert photons to electric charge for digital image production FFF (Flattening filter free) beam (Radiotherapy) See Flattening free filter (FFF) beam The signal produced, stored and subsequently read off each CCD FFT (fast Fourier transform) element is proportional to the light landing on it, with each CCD (General) See Fast Fourier transform (FFT) representing a pixel within an image. The light guide used to focus light photons onto CCD arrays Fibre optics can be formed from lenses or a fibre optic taper (termed a ‘taper’ (General) An optical fibre is a specialised, flexible strand of glass by the fact light from the wide area of the phosphor is tapered F or plastic used to convey light energy between two points. down to the small area of the CCD array). Fibre optic tapers An optical fibre consists of a high refractive index core coated make light collection more efficient than lens arrangements as with a lower refractive index cladding. Due to the difference in the scintillating phosphor is directly coupled to CCD elements. refractive indices between the two materials, light travelling in Specifically, a fibre optic taper is a bundle of single glass fibre the core experiences total internal reflection at the boundary with optic strands which transport light photons through total inter- the cladding and is transmitted along the length of the fibre. This nal reflection. The taper arrangement ensures that the orienta- geometrical optics analysis is no longer valid where the fibre core tion of the individual fibres remains between the phosphor face diameter is on the order of microns and fibre transmission must be and the CCD array. This ensures that the location of an incident explained by reference to the fibre as a waveguide. This analysis x-rays is accurately reproduced when a digital image is formed shows that where the fibre diameter approaches the wavelength (Figure F.12). of the light transmitted, some of the light (the evanescent wave) Related Articles: CCD (Charge-coupled device) actually travels along the fibre through the cladding. Further Readings: Dowsett, D. J. 2006. The Physics of Fibre optics are used primarily for data transmission, where Diagnostic Imaging, Oxford University Press, p. 340; Grant, J. bursts of light transmitted down the fibre convey digital informa- 1998. Fiberoptic coupling increases imaging efficiency and view. tion between an electronic transmitter and receiver. At the trans- LaserFocusWorld. mitter, the data to be transmitted are converted into light pulses, Hyperlinks: www .l aserf ocusw orld. com /fi ber- optic s /art icle/ typically using a miniaturised light emitting diode (LED) or laser 16547 774 /fi bero ptic- coupl ing -i ncrea ses -i magin g -effi cien cy -an d diode. The light is coupled into the fibre for transmission to the -vie w receiver. At the receiver, the pulses are reconverted into electronic signals using a photodiode coupled to the fibre. In general tele- Fibre tracking communications, optical fibres can support higher data rates than (Magnetic Resonance) Fibre tracking is a method for constructing conventional copper wire connections. and visualising axonal tracts based on anisotropic diffusion (e.g. In an optical endoscope, a bundle of optical fibres is used to in cerebral white matter and muscles). See Tractography for convey an image from the distal tip of the endoscope to the eye- further details. piece. For optical image transmission, the fibre bundle must be Related Article: Tractography ‘coherent’, i.e. every constituent fibre must retain the same relative position within a cross section at the distal and proximal ends of FibroScan® the bundle. Fibre bundles are also used to deliver illumination (Ultrasound) The FibroScan® is a specialised ultrasound machine from a source to a distant target. Bundles used for illumination that uses the transient elastography technology to measure liver need not be coherent. shear wave speed (in metres per second) and stiffness (in kilopas- In the MRI environment, fibre optics finds application where cal) at a frequency of 50 Hz. It is a non-invasive, quantitative and the use of conventional conductive wires for data transmission reproducible technique to assess fibrosis (scarring) and steatosis and signal acquisition is limited by RF interference and RF heat- (fatty change) in the liver. ing risks. For example, in a conventional pulse oximeter, a con- Related Article: Transient elastography ductive cable connects the finger clip sensor placed on the patient to the pulse oximeter display. The finger clip houses a LED and Fick’s method photodiode and transmits light through the finger. Analysis of the (Nuclear Medicine) Fick’s method is a technique for determin- light transmitted generates a pulse waveform and a value of % ing consumption of a substance by an organ. It is calculated from oxygen saturation. In an MRI compatible pulse oximeter, fibre the product of the arteriovenous concentration difference of the optics is used to transmit and collect light to and from the finger substance and blood flow. Mass is preserved, and, following that, Fick’s principle 364 Field emission display (FED) FIGURE F.12 X-ray photons incident on phosphor will produce light photons. These will be transported by total internal reflection through fibre- F optic tapers to CCD arrays. Multiple CCD arrays are combined to form a CCD x-ray detector. the mass of a substance entering an organ minus the mass of the One example of their use is to mark anatomy or tumour posi- substance leaving the organ equals the mass of substance retained tion (see for example van der Heide et al. 2007 for gold markers in the organ. to mark prostate position for radiotherapy). Additionally, fiducial Related Article: Fick’s principle markers are used to correlate images taken with different imag- ing techniques (see for example Parker et al. 2003 for markers in Fick’s principle prostate to register CT and MRI images). (Nuclear Medicine) The Fick’s principle states that the rate of Further Readings: Parker, C. C. et al. 2003. Magnetic reso- change of tracer in an organ or tissue is equal to the difference nance imaging in the radiation treatment planning of localized between the amount of tracer arriving (usually via arterial blood) prostate cancer using intra-prostatic fiducial markers for computed and the amount leaving (via venous drainage) the organ or tis- tomography co-registration. Radiother. Oncol. 66(2):217–224; sue. If Ca and Cv denote concentration of tracer in arterial and van der Heide, U. A. et al. 2007. Analysis of fiducial marker-based venous blood, then the amount of tracer Mt in the tissue can be position verification in the external beam radiotherapy of patients described as with prostate cancer. Radiother. Oncol. 82(1):38–45. dM Field coverage t = QCa - QCv - Y dt (Radiotherapy) This term relates to the size of the treatment field being sufficiently large to provide adequate uniform dose cover- where age across the required planning target volume from the beam’s Q is the blood flow eye view. One of the aims of planning is to ensure that a suffi- Y is the rate of loss through excretion ciently high (typically 95%) isodose line completely encompasses the target volume, and so the planner must design a plan to pro- If Y is zero, there is a simple relationship between the rate of vide this. change of Mt and the extraction fraction E (see related article): Related Articles: Target dose distribution, Isodose surface, Planning target volume, Treated volume dMt Ca - ) - = ( - ) ( C = v QCa QCv Q Ca Cv = CaQ = CaQE dt Ca Field echo (Magnetic Resonance) See Gradient echo (GE) Related Article: Extraction fraction Further Reading: Peters, A. M. 1998. Fundamentals of tracer Field emission display (FED) kinetics for radiologists. The Brit. J. Radiol. 71:1116–1129. (Diagnostic Radiology) FED is a form of flat panel display and may be compared to the more common LED and LCD displays FID (free induction decay) found in modern television and computer monitors. (Magnetic Resonance) See Free induction decay (FID) It is based on a two dimensional array of minute cathode ray tubes (CRT), each with an evacuated space containing an Fiducial markers extremely small pointed cold electron emitter, a fine conducting (Radiotherapy) Fiducial markers are markers used as reference grid and a coloured phosphor anode (in colour displays, one CRT points in medical imaging. unit forms a single pixel of one colour). Field emission x-ray tube 365 F ield of view (FOV) Unlike normal CRTs, no heater element is needed to cause the case, the superior and inferior field margins can even be double electrons to be emitted by the cathode. Rather, a positive volt- the standard field margin. age applied to each grid causes electrons to be emitted and pass Abbreviation: PTV = Planning target volume. through, where they are further accelerated towards the phos- Related Articles: Penumbra, Planning target volume (PTV) phor-coated anode by a large positive anode potential, where their energy is converted to visible light. Field patching The non-linear current/voltage property of these mini-emit- (Radiotherapy) See Patch field technique ters allows for easy pixel addressing using a 2D grid of conductors running across and down the screen. Each pixel is selected by Field selection powering the appropriate column and row simultaneously, whilst (Radiotherapy) Field selection is the choice of the number of the brightness of each pixel is controlled by varying the time for treatment fields (beams) and the directions of those fields. Field which the voltage is applied. selection depends on the locations of the tumour and the organs at Despite their requirement for a vacuum within each cell, prob- risk, and the treatment protocol. lems with practical cathodes, and the inefficiency of energy con- version to light, they have been shown to promise much higher Field size power efficiency than present LCD displays. (Radiotherapy) The field size describes the lateral dimensions of They presently remain in the development/prototype stage. the radiation field. Often, it is defined according to the distance Related Article: Cathode ray tube between the two 50% intensity levels in a lateral dose profile Hyperlink: http: / /en. wikip edia. org /w iki /F ield_ emiss ion_ (Figure F.13). display In photon and passive scattering proton therapy, the field size Field emission x-ray tube is defined by the collimator settings. In pencil beam scanning pro- (Diagnostic Radiology) The first x-ray tubes (at the beginning of ton therapy, the field size is defined by the outermost positions of 20th century) used ‘cold emission’ cathodes. These had low elec- the individual proton beams. F tron emissions and were soon replaced by heated cathodes, using Related Articles: Passive scatter proton therapy, Collimator thermioning emission. This method produces higher anode cur- rents and is currently the most widely used one. However it do not Field of view (FOV) have sufficient focusing resulting in relatively large effective focal (Diagnostic Radiology) FOV of a CT scanner refers to the diame- spots. Despite the fact that there are many methods to improve ter of the circular area over which CT data are acquired (scan field this, current

x-ray tubes with heated cathode have technological of view [SFOV]) or over which data are reconstructed (recon- limit of the spatial resolution of the radiograph. struction field of view [RFOV]). Research during the recent decade has used bundles of car- The maximum SFOV of a clinical CT scanner is generally on bon nanotubes (CNT) aiming to produce more effective ‘cold the order of 50 cm (Figure F.14). emission’. With this method the electron emission (i.e. the The RFOV can be equal to, or less than, the SFOV. To achieve anode current) depends solely on the electrical field between an increased resolution, the pixel size can be reduced by decreas- cathode and anode of the x-ray tube and the type of the CNT ing the RFOV. The minimum reconstruction scan field of view is material. The field emission cathodes can produce very fast generally on the order of 5 cm. sequences of x-ray pulses due to the lack of thermal inertia (typical for heated cathodes). Also, they can produce very Field of view (FOV) focussed thin beams of electrons, hence the effective focal spot (Nuclear Medicine) FOV of a scintillation camera system is basi- becomes very small, increasing manifold the spatial resolution cally determined by the size of the scintillation detector and the of the images. The main challenge (at the moment) for these new field emis- sion x-ray tubes is the limited anode current. Experimental mod- els produce between several milliamperes, up to one ampere anode current (at about 150 kV anode voltage). These new tubes have been successfully installed in some low power x-ray medical equipment (e.g. micro CT systems). Further Readings: Zhang, L., J. Lu, Y. Lee, S. Chang and O. Zhou. 2017. Carbon nanotube based field emission X-ray technol- ogy. In Russo, P. (ed.) Handbook of X-ray Imaging: Physics and Technology, CRC Press. Field margin (Radiotherapy) When planning radiotherapy treatments, field margins are often added to each beam’s eye view projection of the planning target volume (PTV), to account for the finite beam penumbra. Typical values would be 0.5 cm; however, the field margin depends on the beam arrangement. For example, in axial coplanar treatments with multiple fields, large field margins are needed for definition of the superior and inferior field borders to avoid under-dosages in the equivalent areas of the PTV, since FIGURE F.13 Field size: typically defined as the distance between the these areas do not have full contribution from other beams. In this two 50% intensity levels in a lateral dose profile. Field strength 366 F ield-effect transistor (FET) Maximum SFOV Head SFOV All detectors active Active detectors FIGURE F.14 Diagram demonstrating (a) maximum SFOV and (b) head SFOV. (Courtesy of ImPACT, UK, www .impactscan .org) depicted image. More contiguous factors that determine the FOV inhomogeneities of the RF coils used in the system. A uniform RF are ring size (PET), collimator design (scintillation camera), edge field is not always required, depending upon the area of the body packing effects (scintillation camera) and the detector surface that is being imaged. Structures close to the surface of the body area (all detector systems). maybe imaged using a surface coil, which is not uniform, while Since parts of the FOV are unsuited for activation, quantifica- imaging of internal structures may use a more uniform RF coil tion or imaging due to a number of reasons (edge packing, etc.), design such as a birdcage coil. it is more relevant to discuss the useful field of view (UFOV) and Abbreviation: RF = Radiofrequency. the central field of view (cFOV). Related Articles: B0 homogeneity, B0 inhomogeneity, B1 Related Articles: Central field of view (cFOV), Useful field of homogeneity, B1 inhomogeneity, RF uniformity F view (UFOV) Field verification Field strength (Radiotherapy) External beam radiotherapy is delivered as a set (Magnetic Resonance) The strength of a field is the magnitude of of treatment fields. It is recommended to verify each field inde- its vector value. Two physical quantities may be referred to as the pendently. This may consist of two stages: verification prior to magnetic field: the magnetic field H and the magnetic induction treatment and on-treatment verification. The verification is done B (also called magnetic flux density). A magnetic field can either by comparison of the shapes and the positions of the planned and be produced by electrical current or by a permanent magnet. The treated fields. magnetic field generated by a current can be calculated from the Field Verification Prior to Treatment: This is often partic- Biot–Savart law or Ampere’s law. When a magnetic field has been ularly useful for complex treatment such as intensity-modulated generated in a medium, the response of the medium is its mag- radiotherapy (IMRT) with dynamically moving multileaf col- netic induction B. The relation between magnetic induction and limator (MLC) leaves. Field shapes may be verified in com- magnetic field is given by the equation B = μH, where μ is the parison with a template. The intensity distribution of an IMRT permeability of the medium (H m−1). Although the term ‘mag- field may be verified using integrated imaging with film or a netic field’ was historically reserved for H, the magnetic induc- digital imaging device such as an electronic portal imaging tion B is now understood to be the more fundamental entity and device (EPID). most modern writers refer B as the magnetic field. The magnetic Field Verification on Treatment: This generally involves induction (or magnetic field) B is measured in Tesla (T). In MRI, imaging the position of the patient’s anatomy relative to the treat- the strength of the main magnetic field varies from 0.2 to 3.0 T in ment field using film or a digital imaging system such as an EPID. clinical practice; whereas, in research, magnets with values of 7 Abbreviations: IMRT = Intensity-modulated radiotherapy, T to higher than 11 T are used. Usually, the main magnetic field MLC = Multileaf collimator and EPID = Electronic portal imag- is generated by superconducting electromagnets. The main mag- ing device. netic field determines the resonance frequency (called the Larmor Related Articles: Radiotherapy verification, Port film, Portal frequency) and the net magnetisation, which is higher for higher radiograph, Electronic portal imaging, Intensity-modulated fields. Furthermore, the T1 relaxation time strongly depends on radiotherapy the magnetic field strength. Related Articles: Larmor frequency, Net magnetisation, Field weight Magnets (Radiotherapy) See Beam weight Field uniformity Field-effect transistor (FET) (Magnetic Resonance) There are two types of field in an MRI (General) Field effect transistors (FET) are widely used for scanner, the main static magnetic field and the RF field. amplification of digital or analogue weak signals, DC switches or The main static magnetic field is referred to as B0 and typi- as oscillators (Figure F.15). cally ranges from 0.1 to 3.0 T for clinical use. The uniformity of In the FET, the current flows along a semiconductor path the magnetic field is an important criterion of the quality of the between the source and the drain electrodes. Its diameter defines magnet as non-uniformities can lead to image artefacts. Magnets FET conductivity. are, therefore, constructed to be as homogeneous as possible, There are two types of FET–junction FET (JFET) and the especially at the isocentre. metal-oxide-semiconductor FET (MOSFET) (Figure F.16). The RF field (also known as B1) uniformity is affected by The wide application of FET in digital radiography (DR) is two main factors, the first is the interaction between the RF due to their good performance in circuits and systems requiring field and the object being imaged and the second is the inherent high impedance. Filament circuit 367 F ilament current time for initiating the flow of the anode current. The time for rapid n-channel p-channel change of the anode current depends on the thermal time-constant of cathode wire and could be on the order of 50–200 ms. In case D D of grid-controlled tube (by the Wehnelt electrode), the filament circuit contains a subcircuit for supplying power to this additional G G electrode. Related Articles: Cathode, Filament, Filament current, Filament heating, High voltage generator, High frequency S S generator Filament current FIGURE F.15 Symbols of field-effect transistors. (Diagnostic Radiology) This is the current which flows through the x-ray tube cathode filament and heats it. This current (If) is usually several Amperes and heats up the cathode to above 2000°C, regulating If through the filament voltage (Uf, the FT output voltage) and affects the number of thermal electrons cre- ated in the filament, hence the tube (anode) current (Ia). The filament current is limited by the maximum permissible temperature of the filament coil. Thus the limiting factor for Ia at low kVp is the temperature of the filament (see the articles on Space-charge effect and Tube load), while the limiting factor for Ia at high kVp is the temperature of the anode. Saturation of the tube current Ia for anode voltages below 50 kV is often called ‘space F charge limited’ operation of the x-ray tube. Note, in Figure F.17, the limitation of the anode current Ia at 50 and 60 kV due to the FIGURE F.16 Symbol of MOSFET (n-channel). fact that all thermal electrons (created by the filament) have been extracted by the accelerating voltage Ua. The maximum filament current is limited to a set value. Its Filament circuit adjustment is made with different kV and mA settings using the (Diagnostic Radiology) This is the electrical circuit which sup- so-called, three-point technique to properly select the necessary plies and controls the filament current If, which determines resistors in the filament circuit (for: low kV at high mA, high kV the temperature of the cathode and, hence, the anode current at low mA, low kV at low mA). Ia during the exposure. The filament circuit is diagrammati- Often the filament emission chart is combined with another cally presented as part of the high voltage generator and high chart showing the relation between the filament voltage, Uf, and frequency generator electrical circuits – see the eponymous filament current, If. This is a linear relationship (often Uf change articles. from 2 to 8 V creates If from 4 to 6 A). The filament current If is supplied to the cathode through the For more details, see the article about Filament circuit. filament transformer (FT) – a step-down transformer with ratio Related Articles: Cathode, Filament circuit, Filament heating between 1:10 and 1:20. It generates a filament current on the order of 3–5 Amperes (A) through the cathode wire. The voltage of the secondary side of the FT is relatively low (around 10–20 V), but it is connected to the cathode which is at very high negative mA 600 potential (kV). Due to this reason, special insulation is necessary between the primary and secondary coil of the FT. One way to achieve such insulation is by winding the primary coil over the core of the FT, placing over it a porcelain cylinder and wind- ing the secondary coil over this cylinder. Additionally, the FT is 400 placed together with the high voltage transformer (which gener- ates the accelerating anode voltage, Ua, kV) in the high voltage 50 kV box with isolating oil. Ia The current through the cathode filament (filament current If ) depends on the maximum permissible power for selected focal spot (Pmax) and the filament circuit assures anode cur- 200 rent (Ia) < P(max)/Ua for each selection of focal spot and anode voltage. In classical high voltage generators, a variable resistor is used to select the set the x-ray tube anode current (Ia, mA), by changing the voltage over the primary coil of the FT. In contemporary high frequency generators, the filament current 0 is controlled by controlling the frequency of the current passing 4 4.5 5 5.5 through the FT. If (A) Normally the filament current keeps the cathode constantly preheated (with approximately 2 A filament current) in order to FIGURE F.17 Filament emission curves (Ia as a function of If with Ua keep short the preparatory period (before the exposure), i.e. short as a parameter). 80 kV 60 kV Filament heating 368 Filament resistor 0.3 0.2 0.1 0 2000 2200 2400 2500 FIGURE F.19 View inside an x-ray tube during fluoroscopic opera- K tion. (Courtesy of A. Litchev and G. Tatarev, Medical University Plovdiv, Plovdiv, Bulgaria.) FIGURE F.18 Emission current (density of electron flow) as

a function of filament temperature. in its focusing cup. Just opposite it (on the left) is the glowing hot rotating anode. F Filament heating Being opposite to the very hot anode (at some 2.5 cm dis- (Diagnostic Radiology) The variation of the anode current (tube tance), the cathode filament is additionally indirectly heated from current) – the thermal electrons flying from cathode to anode – is it (Figure F.19). This can affect the production of the thermal achieved by changing the temperature of the cathode, which in electrons and effectively create a positive feedback, which could turn is achieved by changing the filament current, If (Figure F.18). destroy the x-ray tube (the very hot anode heats the cathode, thus The density of the thermal emission current is described by the increasing the cathode temperature; this increases the anode cur- Richardson equation: rent, which additionally heats the anode, etc.). A special regula- tory circuit with sensor compensates for the influence of the hot J = A T 2e-w / kT 0 0 , anode over the cathode temperature (this is part of the filament circuit). where The filament is heated with current, operating at frequency J0 is the density of the emission current derived from the main frequency. The same frequency is applied T is the temperature of the emitter (in K) to the anode voltage. This may cause a resonance effect and, k and w are constants (k – Boltzmann constant, w – work func- thus, increased ripple. To prevent this phenomenon, modern x-ray tion, for tungsten = 4.5 eV) equipments use an inverter in the filament circuit (which operates A0 is the constant depending of the material of the emitter (for at higher frequency). In this case, pulse-frequency modulation can tungsten = 60 A cm−2 K−2) be used to give very precise control of the filament current. Related Articles: Cathode, Filament circuit, Filament current, The very high temperature of the filament leads to some evap- Tube current, Tube kilovoltage oration of the tungsten wire. This evaporation leads to shortening of the life of the cathode (thinning it). Normally, the cathode life Filament (of an x-ray tube) at this temperature is not more than 1000 working hours. Due (Diagnostic Radiology) The heated tungsten wire of the x-ray to this reason, the cathode is heated to this high temperature for tube cathode produces thermal electrons. The electrical resis- limited time only (during the x-ray exposure). tance of the cathode filament is relatively high and changes from To heat the cathode from room temperature to 2700 K takes approximately 0.1–0.3 Ω when cold, to 2–6 Ω when heated above time; so, in order to keep the heating time short, the cathode stays 2000 K (ohmic heating). For more information, see the article on always preheated at temperature around 1500 K. The preheat- Cathode. ing is achieved by applying a constant stand-by filament current Related Articles: Cathode, Filament current, Filament heat- through the cathode (less than 1 A). This way, the time to heat ing, Focal spot up the filament from the preheating to the requested temperature is much shorter (less than a second). When performing radiog- Filament resistor raphy, the operating x-ray switch normally has two phases (two (Diagnostic Radiology) The filament circuitry of the x-ray gen- steps button). The first stage of the button pressing (known as tube erator includes a number of special resistors which effectively preparation, or prep stage) is associated with heating the cathode control the filament current through the cathode wire, hence filament to the necessary temperature and rising the rate of anode the temperature of the cathode, and thus the anode current. rotation (in the case of a rotational anode x-ray tube). The second Most of these resistors are connected in series with the vari- stage of the button pressing applies the high voltage and produces able resistor used to select the anode current (mA selector) the exposure. and with the FT. The variation of resistance leads to change of Figure F.19 shows the cathode of the x-ray tube during fluo- the input voltage of the FT, hence its output voltage and fila- roscopic operation (all filtration has been intentionally removed). ment current (see the articles on High voltage generator and The heated filament (the small glowing wire on the right) is seen Filament circuit). A/cm2 Fill-in factor 369 Film badge There are sets of resistors which limit the filament current to a set value (depending on the x-ray tube type). These resistors are adjusted by the service engineer as per different kV and mA settings (for: low kV at high mA, high kV at low mA, low kV at low mA). Another set of resistors are associated with the selection of the focal spot (i.e. selection of fine or broad filament wire). These resistors limit the maximal current which can pass through the respective wires. Related Articles: Cathode, High voltage generator, Filament current, Filament heating, Filament circuit, Focal spot Fill-in factor (Diagnostic Radiology) See Detector fill factor Filling factor (Magnetic Resonance) The filling factor is a measure of the interaction of an RF coil with the object being imaged in that coil. It is defined as the ratio of the total magnetic energy in FIGURE F.20 Example of film badge. the transverse component throughout the volume imaged to the total magnetic energy throughout space for a given coil. The energy stored per unit volume in a magnetic field B can be The film is sensitive to x-rays and gamma radiation but also shown to be to light – therefore, the film must be kept in a light-tight holder. F The film consists of the emulsion composed of crystals of silver B2 bromide AgBr (Ag+ and Br−) which is stuck with a gelatine to a Energy per unit volume = polyester base. When the emulsion is irradiated with X or gamma 2m 0 photons, free electrons are created by the photoelectric effect and Compton scattering process. These free electrons remove the The energy stored in the rotating transverse component B1 (i.e. electrons from bromide and neutral atoms of Br are absorbed by the detectable component) in MRI is gelatine. The electrons passing through crystal neutralise silver ions. As a result of the irradiation, a latent image is produced in Energy = ò B1 dV the crystal. To see this image, it is necessary to develop the film 2m0 using a procedure depending on the type of film. After the devel- opment, fixation and hardening the latent image is observed in the The filling factor is then form of dark silver grain speck on the film which is called black- ening. Its quantity is measured by its optical density D defined as ò B1 dV Fillingfactor = 2m0 D = log(I0 /I ) Totalenergy where For a simple coil, the total energy stored is I0 is the incident intensity of visible light I is the transmitted intensity by the developed film 1 Totalenergy = LI 2 2 In Figure F.21, the dependence of the density D on the expo- sure (log E) is shown. It is linear in the region corresponding to where exposure values between E1 and E2. The film may be used to I is the current in the coil determine the dose after calibration using only the linear region. L is the coil inductance To produce in air an optical density of one, the dose of about 2 × 10−4 Gy is required for a typical film. Filling factor clearly influences signal to noise ratio (SNR). The film badge holder usually contains a set of filters (e.g. of Higher filling factors can be achieved by closer conformity of a cadmium to capture neutrons in [n, γ] reaction). The example is given coil to the anatomy of interest (e.g. array coils or flexible shown in Figure F.22. surface coils). Related Articles: Dosimeter, Integrating dosimeter Further Reading: Doty, F. D., G. Entzminger, C. D. Hauck Further Readings: Dendy, P. P. and B. Heaton. 1999. and J. P. Staab. 1999. Practical aspects of birdcage coils. Journal Physics for Diagnostic Radiology, 2nd edn., Institute of Physics of Magnetic Resonance 138:144–154. Publishing, Bristol, UK, pp. 89–90, 323–325; Hobbie, R. K. 1997. Intermediate Physics for Medicine and Biology, 3rd edn., Film badge Springer-Verlag, New York, 421pp; Knoll, G. F. 2000. Radiation (Radiation Protection) Film badges (Figure F.20) are used for Detection and Measurement, 3rd edn., John Wiley & Sons, Inc., personal dosimetry of the staff exposed to radiation, such as those New York, 730pp; Graham, D. T. and P. Cloke. 2003. Principles employed in x-rays and nuclear medicine laboratories as well as of Radiological Physics, 4th edn., Elsevier Science Limited, in radiotherapy units. Edinburgh, London, UK, 337pp. Film base 370 Film fog Converts film density into digital pixel values D3 Saturation Analog/digital converter D2 Scan Film density 620 Light Linearity Array of sensors Laser A pixel in region the digital image D FIGURE F.24 Schematic block diagram of film digitiser. (Courtesy of 1 Sprawls Foundation, www .sprawls .org) Toe 0.58 E1 E Film blackening 2 (Diagnostic Radiology) Film blackening is the result of expo- Exposure sure to a film that is then chemically processed. For radiographic films, the more precise term for film blackening is optical density. FIGURE F.21 Dependence of the optical density of blackening on expo- sure (in log-log scale). Film changer F (Diagnostic Radiology) See Cassette changer Film digitisers (Diagnostic Radiology) Film digitisers are instruments that con- vert images recorded on film into digital images. Conventional scanners used with computers are one type of image digitiser but usually not designed to scan transparent films like radiographs. Specific features of digitisers designed for radiographic film include a large area of coverage (for chest films, etc.) and a rear illumination source for scanning transparencies (Figure F.24). The most important parameter which characterises a film digitiser for radiographic purposes is the digital image pixel size, commonly quantified in dot per inch (DPI). This is related to the requested spatial resolution and the pixel depth, set in number of bits per pixel, which is related to the dynamic range of the grey levels of the pixel. Film dosimetry FIGURE F.22 Example of a set of filters used for film badge. (Diagnostic Radiology) See Film badge Film emulsion Emulsion ~10 μm (Diagnostic Radiology) The emulsion is the active component in which the image is formed and consists of many small sil- Base ~150 μm ver halide crystals suspended in gelatine. The gelatine sup- ports, separates and protects the crystals. The typical emulsion is approximately 10 μm thick. The emulsion is coated onto the Emulsion film base which provides a rigid support. Radiographic film can be made by double layer emulsion in order to increase the sensi- FIGURE F.23 Cross-sectional view of a radiographic film showing the tivity or by single layer emulsion in order to increase the spatial base with an emulsion on both sides. (Courtesy of Sprawls Foundation, resolution. www .sprawls .org) Film fog Film base (Diagnostic Radiology) In conventional radiology, fog is any (Diagnostic Radiology) The base of a typical radiographic film undesirable density or darkness that appears in a film that is not a is made of a clear polyester material about 150 μm thick as illus- result of the exposure forming the image. Sources of radiographic trated in Figure F.23. It provides the physical support for the other film fog include: film components and does not participate in the image-forming process. In some films, the base contains a light blue dye to give • Exposure to other sources of x-ray or gamma radiation the image a more pleasing appearance when illuminated on a • Accidental exposure to light (defective cassettes, light view box. leaks into darkroom, etc.) Optical density Film holder 371 Filter • Overdevelopment (called chemical fog) because of high and ready to be seen on a viewing box. The final visible image developer temperature, excessive development time, contains areas with variable darkness (depending on the concen- incorrect or contaminated chemistry tration of opaque Ag atoms), which is proportional to the inten- • Storage of unexposed film for long times and at high sity/energy of the light or x-ray photons. temperatures Related Articles: Film emulsion, Silver bromide, Latent image, Developer, Fixer, Fixing agent, Acetic acid in film pro- Fog adds to the density in the

‘toe’ region of a film character- cessing, Accelerators in film development, Thiosulphate in film istic curve. That is the region of the curve relating to very low processing, Washing in film processing, Dark room, Automatic or no applied exposure. The optical density value of unexposed film processor, Film transport, Underdevelopment film is due to the intrinsic density of the film base plus any other Further Reading: Dowsett, D. J., P. A. Kenny and R. E. unexpected factors giving rise to film fog. Therefore, the optical Johnston. 1998. The Physics of Diagnostic Imaging, Chapman & density of an area in a film that has not been exposed is designated Hall Medical, London, UK. as the ‘base + fog density’ (the optical density D of such area is normally below 0.2D). Film screen contact Related Article: Characteristic curve (Diagnostic Radiology) Film screen contact refers to the close- ness of contact between the film and the surface of the intensi- Film holder fying screens in a radiographic cassette. If there is space (not (Diagnostic Radiology) Film holder (also known as cassette) is good contact) between a film and screen, it produces blurring and a thin flat light-tight container for the x-ray film (and the respec- reduces visibility of detail. There are established tests for film tive phosphor-intensifying screens). The cassette should keep the screen contact within quality control procedures (Figure F.25). film out of the reach of any light, put the film and screens in very close contact and allow the x-rays to pass through and expose the Film transport film. The front side of the cassette has minimal x-ray absorption. (Diagnostic Radiology) The film transport is the mechanism F One intensifying screen is mounted inside this part of the cas- within a film processor that moves or transports film through the sette. The back side of the cassette has to be only sturdy, as x-ray developer, fixer and wash tanks. It consists of a series of rollers absorption is not more important. This side can also include a thin and guides as shown in Figure F.26. lead sheet to prevent backscatter from the patient table to expose The film transport can be easily removed from the processor the film from the back. Another intensifying screen is mounted for cleaning as shown in Figure F.27. inside this part of the cassette. Usually, this screen is thicker than the entrance screen. The film holder (cassette) is kept locked with special latches; these are different in the case of hand develop- Film type ment or automatic development of the film. Computed radiogra- (Diagnostic Radiology) A film is classified by its characteristics. phy systems use similar-sized holders (but with different inner For radiography, there are different types of films designed for components) to hold the storage phosphor. specific clinical applications. The two major characteristics that Related Articles: Cassette size, Cassette carriage, Screen film are associated with film type are exposure sensitivity (speed) and contact contrast. Further Reading: Thompson, M., M. Hattaway, D. Hall and Related Articles: ASA, Speed of film S. Dowd. 1994. Principles of Imaging Science and Protection, W.B. Saunders Company, Philadelphia, PA. Filter (Nuclear Medicine) This article refers either to the filtering Film processing applied in image processing or physical filtering of particles. (Diagnostic Radiology) Film processing is the procedure of trans- forming the latent film image to a visible one. Classical x-ray radiography relied very much on film processing. This chemical Image blurring process included a number of specific phases: film development, poor film-screen contact followed by fixing, washing and drying. All these are described in detail in their specific articles. In short, the photographic film emulsion contains silver bro- mide (AgBr) and free silver ions moving within its cubic lattice. Each crystal also includes specific impurities (such as sulphur), which form crystal defects – electron traps. When a light (or x-ray) photon excites a bromine atom, it loses an electron. These free electrons are trapped into the crystal defects. The positive free silver ions are attracted into these negative defects, where Film they are neutralised and become Ag atoms (sensitised grains). The combination of areas in the film with different number of sensitised grains forms a latent image. Image During the process of film processing, the emulsion is first developed – a chemical process during which the sensitised grains are stabilised (the exposed AgBr crystals reduced to stable Ag atoms). During the next chemical process of film fixing, the remaining un-sensitised grains (which had not been exposed to FIGURE F.25 Image blur due to poor film-screen contact. (Courtesy of light photons) are removed and washed out. The film is then dried Sprawls Foundation, www .sprawls .org) Filter, compensating 372 Filtered back projection Image Processing: Software filters in medical images are regions with significant absorption variation (chest radiography, used to increase the diagnostic value of the collected images. skull radiography, etc.). The filter used in chest imaging is Trough Example of filters that are used are smoothing filter (to reduce filter with less absorption in the middle (where the image of the noise) or edge enhancement filter (to increase contrast in regions spine will be) and more attenuation at the edges (where the edges with a high signal gradient). of the lungs will be imaged). Such filters are also useful in angiog- Physical Filtering: The process of physical filtering involves raphy (and especially digital absorption angiography [DSA]). The passing a quantity of water or gas through a filter. Examples of filters are also known as wedge filters. For more information, see physical filtration in nuclear medicine are the filtration of colloid the article on Beam restrictor. particles (see the article Filtration) and the filtration of 11CO-gas Related Articles: Attenuation, Beam restrictor, Diaphragm, to prevent 11CO2-breakthrough in 11C production using a Collimator cyclotron. Related Article: Filtration Filtered back projection Further Readings: Bergqvist, L., S.-E. Strand and B. R. R. (Nuclear Medicine) Filtered back projection is a reconstruction Persson. 1982. Particle sizing and biokinetics of interstitial lym- method used to reconstruct 2 dimensional (2D) images into a 3 phoscintigraphic agents. Sem. Nucl. Med. XIII(I):9–19; Clark, J. dimensional (3D) volume. C. and P. D. Buckingham. 1975. Short-Lived Radioactive Gases In emission tomography, imaging projections are acquired at for Clinical Use, Butterworth & Co., London, UK, pp. 227–229. discrete angles αi. These projections are a 2D representation of the source distribution. Using the image reconstruction technique Filter, compensating called back projection, it is possible to get a 3D source distribu- (Diagnostic Radiology) These are metal (most often aluminium) tion. If one accepts a few simplifications, such as the absence of absorption filters used in radiography to obtain images with scattered events, then one row in each projection can be consid- more uniform contrast. They are especially useful in anatomical ered a 1D representation on the source distribution along a line. F The 1D projections from different angles originating from the same position along the trans-axis can be collected in a sinogram. Film path The first row in a sinogram is the 1D projection from angle αI and the second is αi+1 and so forth until all the projection angles are represented. The number of projections acquired determines the number of rows in the sinogram. In simple back projection, each projection is extracted from the sinogram and placed in all rows in an image matrix (the num- ber of columns is equal to the number of columns in the projection and the number of rows can be arbitrarily chosen, but typically they are the same). This matrix is rotated according to the cor- responding projection angle. This procedure is repeated and the rotated image matrices are summed together to produce a recon- structed 3D image. Images reconstructed using simple back pro- jection do not have an adequate spatial resolution. Developer Fixer Wash Dryer Filtered Back Projection: Because of the low image spatial resolution achieved with simple back projection, a number of fil- FIGURE F.26 The transport system for moving film through the film ters can be applied to the individual projections to enhance the processor. (Courtesy of Sprawls Foundation, www .sprawls .org) resolution. Before filtering, the projection data are transformed FIGURE F.27 The film transport rollers removed from the tanks for cleaning. (Courtesy of EMERALD project, www .emerald2 .eu) Filtration, inherent 373 F inger ring dosimeter using the Fourier transform. Using the Fourier transform, spa- Filtration rate tially varying data can be expressed as a series of sine and cosine (Nuclear Medicine) Filtration rate is the rate of fluid filtered functions. In transformed data, the functions with high frequen- through the kidneys per unit time. It is also known as glomerular cies represent sharp edges and low frequencies large image fea- filtration rate (GFR). A deviant filtration rate can be a sign of a tures. A filter that amplifies high-frequency components relative pathological condition and the parameter can be estimated using to low-frequency components will increase the spatial resolution Tc-99m-DTPA and a SPECT camera. but at the same time lower the signal to noise ratio. The projection Abbreviation: SPECT = Single photon emission computed data are then transformed back to the spatial domain using an tomography. inverse Fourier transform. Related Articles: Tc-99m-DTPA, Single photon emission The filtered projections are then used to reconstruct a 3D computed tomography (SPECT) image with simple back projection. Related Article: Pile up effect Filtration, total Further Reading: Cherry, S. R., J. A. Sorensen and M. E. (Diagnostic Radiology) The total filtration of an x-ray beam aims Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, to reduce the unnecessary low energy x-ray photons. It is pro- Philadelphia, PA, pp. 283–291. duced by a combination of two filter components – inherent and added. Filtration, inherent The inherent filtration consists of existing components of the (Diagnostic Radiology) See Inherent filtration x-ray system through which the x-ray beam passes (tube window, oil in tube housing, beam locator mirror, etc). Filtration (of colloidal particles) The added filtration of the x-ray beam comprises of sheets of (Nuclear Medicine) Filtration is a method for physical quality metal that are added to provide the required total filtration. Most control of colloidal particles, by the use of membrane (ultra) fil- often for mammography this filtration is from molybdenum or tration or gel filtration. In both cases, the measurable quantity rhodium, while for general radiography/fluoroscopy it is made of F is the size of the particle, which is obtained by measurement of aluminium (although some specific metal are also used in specific either the activity of the sample or the number of particles as a examinations – e.g. tantalum, cerium, etc) function of their sizes. In general, the activity of a particle is pro- Total filtration is monitored by international and national reg- portional to the volume or surface of the particle. Thus, larger ulations – for example all x-ray equipment capable of producing particles have normally considerably higher activity than smaller energies above 100 kV, must have total filtration at least 2.5 mm particles. It should be remembered that the activity distribution Al equivalent. obtained does not give the size distribution. Related Articles: Added filtration, Inherent filtration, Practically, small samples of the colloids are passed through Mammography x-ray tube, Tantalum filter polycarbonate filters (Nuclepore® or Millipore®). The pore sizes of the filters are available between 15 and 1000 nm. The filter is rinsed after the passage of the colloid (0.01–0.25 mL sample) Finger ring dosimeter with 1–4 mL distilled water. The filter and the filtrate are then (Radiation Protection) Finger ring dosimeters (Figures F.28 measured for the activity. The activity in the filtrate is given as and F.29) are made from the thermoluminescent material (e.g. percentage of the total activity. LiF:Mg Ti). They are used as an integrating dosimeter to mea- Physical control by the use of filtration of a new radiophar- sure the radiation dose to the fingers of staff, mainly those in maceutical, primary regarding particle size, concentration and interventional radiology and staff working in nuclear medicine stability may be important prior to functional investigations, for laboratories. example of the reticuloendothelial system (RES). The character- istic of the particles is very important for the uptake rate and the distribution in the tissue. Different techniques for particle characterisation have been published by Bergqvist et al. Membrane filtration is also the most common method for sterilisation of radiopharmaceuticals, using

a Millipore filter for removal of various organisms. A common membrane filter size is 0.45 μm, but a pore size of 0.22 μm is used for the sterilisa- tion of products of the blood and preparations containing smaller microorganisms. Related Articles: Biological purity, Quality control, Tc-99m-labelled microcolloids, Tc-99m-labelled nanocolloids, Tc-99m-albumin microcolloid, Tc-99m-albumin microspheres, Tc-99m-albumin nanocolloid, Tc-99m-rhenium sulphide colloid Further Readings: Bergqvist, L., S.-E. Strand and B. R. R. Persson. 1982. Particle sizing and biokinetics of intersti- tial lymphoscintigraphic agents. Sem. Nucl. Med. XIII(I):9–19; Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American Pharmacists Association, Washington, DC; Saha, G. B. 2004. FIGURE F.28 TL finger ring dosimeter in an adjustable holder. (Photo Fundamentals of Nuclear Pharmacy, 5th edn., Springer, New courtesy of Maciej Budzanowski, PhD, Institute of Nuclear Physics, York. Polish Academy of Sciences, Krakow, Poland.) FISP (fast imaging with steady state precession) 374 Fixing agent There are two types of FFAG designs: scaling and non-scal- ing. For scaling FFAGs, the magnetic field in the y-axis (increas- ing distance from the centre of the accelerator ring) follows Equation F.1: k æ r ö By = B0 ç . ) è r ÷ (F 1 0 ø where k is the field index, r is the radial coordinate and B is the magnetic field strength. Non-scaling relaxes the scaling requirement in Equation F.1. Examples of non-scaling FFAG designs include EMMA (Electron Machine for Many Applications) (Machida et al., 2012) and PAMELA (particle accelerator for medical applications) (Peach et al., 2013). Other FFAG designs include NORMA (normal- conducting racetrack FFAG) (Tygier et al., 2017) and RACCAM FIGURE F.29 (Ohmori et al., 2008). TL ring dosimeter worn on the finger. (Photo courtesy of Maciej Budzanowski, PhD, Institute of Nuclear Physics, Polish Academy FFAGs have been proposed for hadron therapy use for their of Sciences, Krakow, Poland.) compact sizes, high repetition, larger energy range, and cheaper price. Related Articles: Synchrotron, Cyclotron F Further Readings: Machida, S. et al. 2012. Acceleration in Typical range of response: 0.1 mSv–1 Sv for the radiation of the linear non-scaling fixed-field alternating-gradient accelerator energy of 15 keV–3 MeV. The dosimeters can be used to integrate EMMA. Nature Physics 8(3):243; Ohmori, C. et al. 2008. High doses over periods of time from 1 to 3 months. field gradient RF system for a spiral FFAG, RACCAM. European Abbreviations: TL = Thermoluminescence and TLD = Particle Accelerator Conference EPAC'08. Joint Accelerator Thermoluminescent dosimeter. Conferences Website; Peach, K. J. et al. 2013. Conceptual design Related Articles: Integrating dosimeter, Thermoluminescent of a nonscaling fixed field alternating gradient accelerator for dosimeter (TLD) protons and carbon ions for charged particle therapy. Physical Further Reading: Dendy, P. P. and B. Heaton. 1999. Physics for Review Accelerators and Beams 16(3):030101; Symon, K. R. et Diagnostic Radiology, 2nd edn., Institute of Physics Publishing, al. 1956. Fixed-field alternating-gradient particle accelerators. Bristol, UK, p. 322. Physical Review 103(6):1837; Tygier, S. et al. 2017. Medical therapy and imaging fixed-field alternating-gradient accelerator FISP (fast imaging with steady state precession) with realistic magnets. Physical Review Accelerators and Beams (Magnetic Resonance) See Fast imaging with steady state preces- 20(10):104702. sion (FISP) Fixer Fixation (Diagnostic Radiology) After leaving the developer, the film is (Radiotherapy) See Immobilisation transported into a second tank, which contains the fixer solution. The fixer is a mixture of several chemicals that perform the fol- Fixed aperture beam restrictors lowing functions: (Diagnostic Radiology) See Beam restrictor Neutraliser: When a film is removed from the developer solu- tion, the development continues because the solution was soaked Fixed field alternating gradient (FFAG) accelerator up by the emulsion. It is necessary to stop this action to prevent (Radiotherapy) Fixed field alternating gradient accelerators, orig- overdevelopment and fogging of the film. Acetic acid is in the inally proposed in the 1950s, also known as FFAGs, are a type of fixer solution for this purpose. circular accelerators that use magnets which are constant in time Clearing: The fixer solution also clears the undeveloped silver (i.e. fixed field) and can accommodate particle orbits of different halide grains from the film. Ammonium or sodium thiosulphate energies (i.e. alternating gradient) (Symon et al., 1956). is used for this purpose. The unexposed grains leave the film Compared to synchrotrons, FFAGs are similar in the way that and dissolve in the fixer solution. The silver that accumulates in both accelerate particles in circular orbits along an annular ring the fixer during the clearing activity can be recovered; the usual of accelerator components (e.g. quadruples, dipoles, radiofre- method is to electroplate it onto a metallic surface within the sil- quency chambers). The difference is that synchrotrons use pulsed ver recovery unit. field to keep particles of different energies within orbits of simi- Preservative: Sodium sulphite is used in the fixer as a lar radii. FFAG relaxes the requirement such that higher-energy preservative. particles circulate along orbits with larger radii than lower ener- Hardener: Aluminium chloride is typically used as a hardener. gies. This means FFAGs have a larger momentum acceptance, Its primary function is to shrink and harden the emulsion. higher beam intensity and simpler operation. Compared to Related Article: Film processing cyclotrons, FFAGs are similar as both use magnetic fields that are constant in time. A main difference is that cyclotrons output Fixing agent constant energy, whereas FFAGs are able to produce a spectrum (Diagnostic Radiology) After leaving the developer, the film is of particle energies. transported into a second tank, which contains the fixing agent FLAIR (fluid attenuated inversion recovery) 375 F lat panel detector solution. The fixing agent is a mixture of several chemicals that usually appearing black or dark grey. It can be used during equip- perform the following functions (see the article on Fixer). ment evaluation and quality control to calculate the quantum and Related Article: Fixer system noise of the x-ray imaging equipment. A flat field image must be taken to determine the noise power spectrum (NPS) and FLAIR (fluid attenuated inversion recovery) thus the noise equivalent quanta (NEQ) and detective quantum (Magnetic Resonance) See Fluid attenuated inversion recovery efficiency (DQE) of the system. (FLAIR) Related Articles: Detective quantum efficiency (DQE), Noise equivalent quanta (NEQ), Noise power spectrum (NPS), Signal to FLASH (fast low angle shot) noise ratio (SNR) (Magnetic Resonance) See Fast low angle shot (FLASH) Flat panel array FLASH (Diagnostic Radiology) See Flat panel detector (Radiotherapy) FLASH radiotherapy involves the delivery of dose at ultra-fast dose-rates, typically >40 Gy s−1. It is believed Flat panel detector that such dose-rates might spare normal tissue while maintaining (Diagnostic Radiology) Flat panel detectors are used as digital tumour control. The mechanisms underpinning FLASH are yet radiographic (x-ray) detectors and can be divided into two main to be elucidated, but it is hypothesised that FLASH tissue sparing subgroups: direct and indirect detectors, Figure F.31. They are stems from the rapid depletion of oxygen. FLASH is not an acro- nym, rather a name stemming from the word ‘flash’ and its classic definition: to move or pass very quickly. Direct detection Indirect detection Related Articles: Hypoxia Further Readings: Esplen N.M., M.S. Mendonca and M. Bazalova-Carter. 2020. Physics and biology of ultrahigh dose-rate F (FLASH) radiotherapy: a topical review. Phys Med Biol. 2020 Jul X-rays X-rays 28. doi: 10.1088/1361-6560/abaa28. Epub ahead of print. PMID: 32721941. Scintillator Flash artefact (Ultrasound) A rapid shift in the Doppler signal will cause a sud- X-ray photoconductor X-rays Visible light den increase in its value. When imaging using colour Doppler, this may manifest as a sudden ‘flash’ of colour on the display screen. This Doppler shift can be caused by sudden movement of the transducer (by the operator) or the patient, or physiological reasons X-rays Electrical charge Photodiode such as cardiac and bowel motions or breathing. Flash artefacts are usually considered ‘false’ signals and therefore some systems have Visible light Electrical charge attempted to compensate for them by use of a ‘flash filter’. Shown below are two images: on the left is a B-mode image TFT- TFT- of a kidney and on the right is the corresponding dual-mode Active matrix array Active matrix array (B-mode plus Doppler) image of the same. The red and blue pixels represent a flash artefact (Figure F.30). – Signal readout – Signal readout Flat field image FIGURE F.31 A schematic diagram of a direct and indirect flat panel (Diagnostic Radiology) A flat field image is an x-ray image taken detector. (Redrawn and adapted from Kotter, E. and M. Langer., Eur. with no object in the beam and is thus, a low contrast image, Radiol., 12, 2562, 2002.) FIGURE F.30 Flash artefact (right). Flattening filter 376 Flattening filter usually of slim design to allow them to be used in conjunction Multiplexer ADC with the traditional Bucky tables and have dimensions of at least 35 × 43 cm2 for general radiography. Charge amplifier They are typically constructed from a large area, active matrix array which is electronically coupled to a solid-state detector, Charge collector either a photoconductor or a phosphor screen. Direct detectors electrode convert the incident x-rays directly to electrical charge by the use of a photoconductor, such as amorphous selenium. Selenium (Z = 34), usually used in direct detectors, has a higher atomic number than silicon (Z = 14), and, therefore, its attenuation coefficient is higher. In contrast, indirect detectors convert the incident x-rays to electrical charge by a two-stage process. Firstly, a scintillation Switch, phosphor screen, such as caesium iodide doped with thallium (or diode or TFT gadolinium oxide sulphide), converts the incident x-rays into vis- ible light wavelength photons and then a photodetector, such as an amorphous silicon photodiode array, converts the visible light photon to electrical charge. Some radiographic systems including Gate line charge-coupled devices (CCD) are sometimes also termed indi- rect flat panel detectors as they use a large-area scintillation phos- phor screen to convert the incident x-rays to visible light photons; Data line but, as the released light photons are then typically focused onto a smaller CCD integrated chip, these are not strictly considered FIGURE F.32 The basic construction of an active matrix array. F to be flat-panel, large-area devices and do not use active matrix (Redrawn and adapted from Kotter, E. and M. Langer., Eur. Radiol., 12, arrays for the readout process. 2562, 2002.) Excluding CCD-based devices, the readout technology used for both direct and indirect detectors is based upon technology designed for LCD monitors and thin film technology. To display Further Readings: Beutel, J., H. L. Kundel and R. L. Van an image on an LCD monitor, a large two-dimensional array of Metter. 2000. Handbook of Medical Imaging: Physics and thin film transistors (TFTs) arranged on a glass substrate is used to Psychophysics, Vol. 1. SPIE, Bellingham, WA; Kotter, E. and charge and consequently illuminate individual pixels to achieve the M. Langer. 2002. Digital Radiography with large-area flat-panel desired intensity. To address each pixel element individually, instead detectors. Eur. Radiol. 12:2562–2570. of wiring each element separately, they are connected together by a series of horizontal and vertical lines, and thus, by using the correct Flattening filter logic, each individual pixel can be accessed and addressed. This (Radiotherapy) The flattening filter is located within the head of a type of array is called an active matrix array. The active matrix linear accelerator (just below the primary collimator) and is used array used in flat panel detectors is an integrated circuit formed out to produce a ‘flat’ beam profile for x-ray photon radiotherapy of a large number of photodetector elements connected to TFTs. It treatment. The flattening filter is typically made from aluminium, can be produced as a large area matrix (currently available 43 × 43 and different flattening filters are used for different beam energies cm2). The active matrix array for a flat panel detector works in an and can be replaced in the beam path by scattering foils whenever opposite fashion to an LCD monitor. The signal for each detected an electron beam has to be produced. incident x-ray photon is stored as charge in each pixel element, and The beam produced from bombarding the target with elec- the array is then used to read out this charge from sequential pixel trons results in a very strongly forward peaked beam. By placing elements, row-by-row, in an active matrix readout. a flattening filter in the beam path,

this peaked profile is modified Figure F.32 shows a typical array used in a medical imaging to a more useable flat profile as illustrated in Figure F.33. It is flat panel detector. Within the array, each pixel consists of the important that the beam steering and angle of impact on the flat- switching element and an element to detect incoming photons and tening filter is correct; otherwise, a sloping profile will result as store them as charge. The image read-out process is controlled by illustrated in Figure F.34. altering the voltage applied across the switching element. Firstly, Therefore, it is clear that the position of the flattening filter to allow each pixel to detect a signal during exposure, the voltage with respect to the beam central axis is critical, and any move- across each switching element is set to an ionisation or ‘off’ state. ment of either will produce a significantly different profile which The signal is then read out by changing the switching voltage row- may be asymmetrical and no longer flat. It is also important that by-row to the conducting or ‘on’ state which allows the charge the energy of the beam is monitored as changes can cause the stored in each pixel to be drained by the charge collector electrode beam to be either over- or under-flattened. The beam uniformity and passed to the multiplexer. The voltage change is controlled by is measured within the head of a linac by the monitor chamber. the gate line driver. As the read-out process is controlled by the Those linacs that operate with more than one x-ray energy external circuitry, each row of pixels requires a separate control will also have more than one flattening filter, and there will be an line driver to alter the switching voltage, and each column its own interlock system in place to avoid the incorrect filter being used amplifier. This process is called the active matrix read-out. Related Articles: Optimal incident beam profiles, Linear Abbreviations: TFT = Thin film transistor and CCD = Charged accelerator, Treatment head, Beam flatness, Beam symmetry, coupled device. Monitor chamber Related Articles: Direct x-ray detection, Amorphous sele- Further Reading: Greene, D. and P. C. Williams. 1997. Linear nium, Active matrix array, Thin film technology, LCD monitors, Accelerators for Radiation Therapy, 2nd edn., IOP Publishing, Silicon diode detector Bristol, UK. Gate line driver Flattening filter free (FFF) beam 377 Flattening filter free (FFF) beam 115 115 110 110 105 105 100 100 95 95 90 90 85 85 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 –200 0 200 –200 0 200 Y (mm) X (mm) (a) (b) FIGURE F.33 Beam profiles measured (a) without and (b) with a flattening filter in place. F dose at a certain depth in a homogeneous material. The flattening Electrons Electrons hit filter is located between the primary collimator and the monitor hit middle target at an chamber. Flattening filters usually have a conical shape and are of target angle composed of high-Z materials and are several centimetres thick in the centre of the cone to produce a flat beam profile. Consequently, the beam dose rate is reduced and the number of scatter photons Flicker is highly increased. General Radiotherapy treatment techniques such as stereotactic radio- therapy (SRT), where inhomogeneous dose distributions are applied, or static and rotational intensity modulated radiotherapy (IMRT), where varying fluence pattern across the beam are deliv- ered, require the use of beams of photons formed without using homogeniser filters. Flat profile Un-flat profile This operating mode is called flattening filter free (FFF) mode. When removing the flattening filter, the electron beam remains Electrons Electrons hit unaffected through the bending magnet system, if present, as well hit middle target at an as in the target and the primary collimator apparatus. The latter is of target offset position used to eliminate parts of the bremsstrahlung photons which are emitted at large angles. The main differences between flattened and unflattened beams arise at the level of the monitor chamber, which deter- mines the monitor units and keeps beam flatness, symmetry and uniformity under control.The bremsstrahlung production target is commonly put on a low-Z material for cooling pur- poses and for reducing electrons produced by the thin target, which have a rather broad angular distribution. The flattening Flat profile Un-flat profile filter also produces electrons which usually reach the monitor chamber. If the flattening filter is removed, the electron fluence FIGURE F.34 An illustration of the impact of beam steering and angle at the level of the monitor chamber is different. The main neu- of impact to the flattening filter on the beam profile. tron sources are the primary collimator, target, flattening filter, jaws and the multileaf collimators. All these components are vendor specific. Flattening filter free (FFF) beam On removal of the flattening filter, the neutron flux can be (Radiotherapy) The bremsstrahlung distribution from high negligible for energies between 8 and 10 MV. For photon beams energy photons is forward peaked and has an energy and inten- energies higher than 10 MV, the neutron component increases and sity variation of the primary photon fluence with the emission must be evaluated taking into account that the generation of neu- angle. trons does not depend only on beam energy. To compensate for this effect flattening filters have been intro- Related Articles: Stereotactic radiotherapy, Intensity modu- duced in the linac treatment head, resulting in an almost uniform lated radiotherapy Flicker 378 Flood field image Flicker The name ‘float’ is given because the representation allows the (General) Flicker refers to the visual ‘flickering’ effect seen on decimal point to ‘float’ anywhere relative to the significant dig- video displays. It is particularly evident on computer displays its. The stored number would consist of a signed digit of a given based on the CRT. length (the significand), and a signed integer exponent, which The display screen of a CRT device is a glass screen coated modifies the magnitude of the number. on the inside with a phosphorescent material. An electron gun For example, in decimal notation, the number 4562.907 is rep- provides a focused beam of electrons which strike the phosphor resented as the significand 4.562907, together with an exponent at a high velocity and cause it to give off phosphorescent light. of 3. Floating point can be used with different bases other than The lifetime of the light emission is called the persistence, and 10 (decimal). this varies depending on the phosphor material. If the persistence Equation F.2 is a symbolic representation of floating point rep- time is short, the image will fade and will have to be refreshed. resentation in decimal notation: Flicker occurs if the refresh rate is low. A refresh rate of 60 Hz, for example may exhibit flicker, whereas a rate of 70 Hz or higher Value Significand exponent will usually result in flicker-free viewing. = ´10 (F.2) e.g. 4562.907 = 4.562907 × 103 Flip angle Because of the flexible positioning of the decimal point, float- (Magnetic Resonance) The flip angle (α) describes the net rotation ing point representation can achieve a much wider range of values of the bulk magnetisation vector by an RF pulse. At resonance, it for the same storage requirements. However although float values is determined by the time integral of B1: are generally of high precision, they cannot faithfully reproduce a real irrational number, and this limiting precision should always a = gòB1 (t )dt be considered in calculations and storage. F For an excitation pulse, it is given with respect to the direction Flood field of the static magnetic field (Figure F.35). The remaining longitu- (Nuclear Medicine) Flood field images are acquired when the dinal component influences the T1-weighting in steady state. For detector is irradiated by a parallel beam of photons with a per- a refocusing pulse, the flip angle influences the echo amplitude. pendicular angle of incidence covering the entire detector surface. Related Article: B1 Flood field images are used to measure image non-unifor- mity in scintillation cameras. Uniformity is the camera’s ability to depict a uniformly distributed field of photons. Camera uni- Float value formity is controlled on a regular basis as a part of the camera (General) In computing, numbers are stored in a representa- quality control process. If the measurements show signs of non- tion that depends on the storage capacity of the system and the uniformity, mathematical corrections should be applied. The uni- required accuracy of the number. A number is generally encoded formity can be measured in two ways: the intrinsic and extrinsic in a string of digits, which can be decoded if the computing sys- uniformity. Intrinsic flood field images are acquired without the tem has information on the coding representation. Float value, or use of a collimator. The source depicted is usually a point source floating point representation, is one such method that allows stor- placed far away from the detector (typically four to five times the age of high precision values over a wide range. detector diameter). Such experimental set up will guarantee a For example, floating point would be a better representation to near uniform radiation of the detector. store the number (p), an irrational number whose decimal repre- Extrinsic flood field imaging is performed using a collima- sentation hence never ends, than the integer representation. tor and a uniform flood source placed as close to the collimator as possible. The area of the flood source must cover the detector area. The information acquired in the uniformity measurements α are then used to update the uniformity correction used for patient M examinations. o Related Article: Scintillation camera Further Reading: Cherry, S. R., J. A. Sorensen and M. E. Phelps. 2003. Physics in Nuclear Medicine, 3rd edn., Saunders, Mz Philadelphia, PA, pp. 234–236. Flood field image (Nuclear Medicine) Flood field image refers to an image used to quantify the uniformity of the detector response. According to National Electrical Manufacturers Association (NEMA) stan- dards, a (intrinsic) flood field image for a scintillation camera should be acquired using 99mTc in a source holder (to prevent side and backscatter) at a distance greater than five times the Mxy detector diameters. The detector surface should not be colli- mated. The count rate should not exceed 20,000 counts per sec- FIGURE F.35 The flip angle α (Mo initial longitudinal magnetisation, ond for small FOV detector. For high count rates, the detector Mz longitudinal magnetisation, Mxy transverse magnetisation in the x–y surface can be shielded by a lead mask with homogenous thick- plane). ness. Image acquisition should be continuous until the central Flow compensation 379 Flow encoding element has registered 10,000 events per pixel. Before evaluation, the other hand, if the repetition time is not too long, the image should be convolved with a smoothing filter. To read the inflow of ‘fresh’, non-saturated spins during the more about evaluation of the flood field image, please see the repetition time will increase the magnetisation com- article Uniformity. pared with the static, saturated, spins and thus give an Related Articles: Scintillation camera, Uniformity increase in signal. These competing mechanisms make Further Reading: National Electrical Manufacturers the modulus signal behaviour versus flow velocity Association. 2007. NEMA Standards Publication NU 1. National biphasic in such sequences. If only one RF pulse is used Electrical Manufacturers Association, Rosslyn, VA, pp. 9–10. per repetition time, as in gradient echo sequences, the signal decrease due to wash-out will not be observed. Flow compensation (See also Inflow effect and Flow void.) (Magnetic Resonance) During an MR sequence, the object 2. Phase effects due to macroscopic flow imaged is expected to be at rest. If this is not the case, and the The movement of spins affects the voxel magneti- object moves during and/or between data acquisition, artefacts sation phase angle in quite a different way than the such as mis-positioning and blurring may occur in the recon- corresponding amplitude. The underlying mecha- structed images. nism has been known for some time and involves A moving spin will accumulate a phase offset when exposed motion of spins along a magnetic field gradient. The to a gradient. This phase is different from the phase of a static

phase shift caused by linear movement during a time spin (see Phase contrast and Velocity mapping). The offset phase (t) along a constant magnetic field gradient (G) in the angle results in mis-positioning and blurring of the object in the absence of RF pulses is proportional to the velocity. Fourier reconstruction (Figure F.36). Movements of higher order, i.e. accelerations, also Flow compensation can be performed by introducing addi- affect the phase angle, but if these effects can often tional gradient lobes prior to the echo readout. The aim of these be disregarded, the voxel phase angle is a sensitive lobes is to cancel out the phase offset induced by motion. and reasonably good measure of linear flow veloci- ties independent of relaxation parameters. (See also F Flow compensation is normally used when large vessels are in proximity to the desired object in the image. A consequence Phase mapping.) of flow compensation is longer TE, as more gradients need to be 3. Pulsation effects introduced prior the echo read-out. Contrary to steady flow, which will not interfere with Compensation both for velocity and for higher order move- the spatial phase encoding, a variable flow resulting in ment such as acceleration and jerk can be done (see the article a varying phase shift during the measurement generally Acceleration compensation). leads to artefacts or mis-positioning along the phase Related Articles: Acceleration compensation, Phase contrast, encoding direction. If the flow is pulsative and no physi- Velocity mapping ological triggering or gating is applied, typical artefacts appear in regular patterns. Hence, studies of pulsative Flow effects flow as well as examinations where a suppression of (Magnetic Resonance) Many types of flow effects (effects on sig- such artefacts is essential should involve a trigging/gat- nal in MR images) are known. Among the most prominent are as ing procedure. follows: 4. Phase dispersion effects If spins within a voxel have a distribution of veloci- 1. Wash-in/wash-out phenomena ties along a certain gradient direction, a range of dif- Sequences using two or more spatially selective RF ferent phases results. This phase dispersion in the voxel pulses, such as the SE and IR sequences, are sensitive reduces the net magnetisation vector amplitude but the to the transport of spins into and out of the excited slice. net phase angle still reflects the net flow. If the dephas- If the flow is coherent over at least a number of voxels ing effects are strong, however, and possibly combined (macroscopic flow) and has components perpendicular with outflow effects, the signal may be destroyed com- to the imaging slice, the signal will decrease due to the pletely (see also Flow void). outflow of spins during the time between RF pulses. On 5. Attenuation due to microscopic flow Similar to phase dispersion effects, incoherent intra- voxel motion (diffusion, perfusion) in a non-homo- geneous magnetic field can result in detectable phase G dispersion, and hence to a corresponding reduction in amplitude. These effects are studied with dedicated MR techniques (see also Diffusion imaging and Perfusion τ imaging). (a) Related Articles: Diffusion imaging, Flow void, Inflow effects, Perfusion imaging, Phase mapping, Velocity encoding, Phase offset Spin with constant velocity Flow quantification Static spin Time Flow encoding (b) (Magnetic Resonance) Flow encoding is also known as velocity encoding. For further information, see the article Velocity FIGURE F.36 (a) Gradients and (b) corresponding phase for a static and encoding (VENC). a spin with constant velocity, respectively. Related Article: Velocity encoding (VENC) Flow imaging 380 F low imaging Flow imaging • Continuous wave (CW). Non-depth-specific ultrasound (Magnetic Resonance) A large number of methods for in vivo device that gives an audio output of the Doppler fre- quantification of blood flow (flow imaging) using MRI have been quencies. A spectral analyser can be added to give a proposed during the last three decades. Among the first scientists sonogram display of the time varying distribution of the to explore the field were Herfkens et al. (1981), who stated that the Doppler frequencies (Figure F.37). information in a conventional MRI was flow dependent; Grant and • Colour flow imaging (CFI) uses pulsed wave Doppler Back (1982), who showed wash-out, or flow void, effects in tubes, to produce a colour-coded map of Doppler shifts in an and Crooks et al. (1983), who published a signal versus veloc- area of an image superimposed on B-mode image. ity curve obtained with a spin-echo sequence. Singer and Crooks • Power Doppler. Similar to CFI, power Doppler uses the (1983) as well as Wehrli (1985) proposed quantitative methods for amplitude of the CFI process output to display the pres- velocity measurements in vessels, based upon wash-in/wash-out ence of Doppler shifts and variation of the amount of effects. Magnetic tagging methods, often known as bolus track- moving scatterers in the image. ing methods have been proposed for use in MRI by, for example • Pulsed wave (PW) spectral Doppler. PW systems per- Shimizu (1986). mit detailed examination of the circulation at a particu- With respect to flow-induced phase effects on motion, the lar depth along a beam. In duplex ultrasound systems, attenuation in modulus images due to loss of phase coherence they can be used with the B-mode and CFI image to within the voxel was used by Waluch and Bradley (1984), who produce a sonogram from a specific sample volume pointed out the clinical possibilities of the so-called even echo within the image (Figure F.38). rephasing effect. Development along another line utilised motion-induced phase effects more directly. Here, Moran (1982) proposed a method for the creation of velocity images by the addition of F flow-sensitive or flow-encoding gradients to a standard pulse sequence. The technique was rapidly developed by, among oth- ers, van Dijk (1984) and Nayler (1986), leading to the today well-established method known as velocity mapping or phase mapping technique. As a result of the work outlined earlier, quantitative flow measurement sequences and evaluation tools have become available on standard MRI scanners, providing the possibil- ity not only to measure velocity and flow, but also to evaluate important physiological parameters, such as shear stress and compliance. Further Readings: Crooks, L., P. Sheldon, L. Kaufman and W. Rowan. 1982. Quantification or obstructions in ves- sels by nuclear magnetic resonance (NMR). IEEE Trans. Nucl. Sci. 29(3):1181; Grant, J. P. and C. Back. 1982. NMR rheoto- mography. feasibility and clinical potential. Med. Phys. 9:188; Herfkens, R. et al. 1981. Nuclear magnetic resonance imaging FIGURE F.37 Continuous wave Doppler with spectral display showing the time-changing distribution of Doppler frequencies. of the abnormal live rat and correlations with tissue charac- teristics. Radiology 141:211; Moran, P. R. 1982. A flow veloc- ity zeugmatographic interlace for NMR imaging in humans. Magn. Reson. Imaging 1:197; Nayler, G. L., D. N. Firmin and D. B. Longmore. 1986. Blood flow imaging by cine magnetic resonance. J. Comput. Assist. Tomogr. 10:715; Shimizu, K. et al. 1986. Visualization of moving fluid. Quantitative analysis of blood flow velocity using MR imaging. Radiology 159:195; Singer, J. R. and L. E. Crooks. 1983. Nuclear magnetic reso- nance blood flow measurement in the human brain. Science 221:654; van Dijk, P. 1984. Direct cardiac NMR imaging of heart wall and blood flow velocity. J. Comput. Assist. Tomogr. 8:429; Waluch, V. and W. G. Bradley. 1984. NMR even echo rephasing in slow laminar flow. J. Comput. Assist. Tomogr. 8:594; Wehrli, F. W., A. Shimakawa, J. R. MacFall, L. Axel and W. Perman. 1985. MR imaging of venous and arterial flow by a selective saturation-recovery spin echo (SSRSE) method. J. Comput. Assist. Tomogr. 9:537. Flow imaging (Ultrasound) In ultrasound, the term flow imaging covers a range FIGURE F.38 Flow imaging in a colour flow scanner. Colour shows of techniques to image and measure flow and flow parameters by the presence of flow in a renal artery and pulsed wave Doppler is used what are generally regarded as Doppler methods. to produce a sonogram from a specific sample volume identified in the The most commonly used techniques are as follows: image. Flow phantom 381 Flow void Other techniques used include ss inversion • Time domain colour flow imaging, similar to CFI auto- correlation techniques. • Colour M-mode displays, showing the time-varying dis- tribution of velocity vectors along a specific scan line. TI • Contrast agent wash-out and wash-in. The uptake of contrast agents and measurement of the intensity of contrast agent density have been used as a measure of perfusion and transit time. (a) Flow phantom ns inversion (Ultrasound) A phantom using fluid flow to test the Doppler and colour flow characteristics of a Doppler ultrasound scanner or device. See Doppler phantom TI Flow quantification (Magnetic Resonance) Two different methods exist for flow quan- tification with MR. Modulus-based techniques rely on the change in the signal (b) magnitude which occurs when moving spins travels out of or into Uninverted tissue magnetisation Uninverted blood magnetisation the imaging slice. With suitable models, a quantification of the Inverted tissue magnetisation Inverted blood magnetisation flow is possible. For more information, see Inflow-effect and Flow F void. FIGURE F.39 The FAIR concept: magnetisation directly after the inver- Phase-contrast techniques are more common and use the sig- sion pulse (left column) and at the time of read-out (right column). (a) nal phase for flow quantification. These methods use a bipolar Slice-selective experiment. (b) Non-selective experiment. gradient shape in the pulse sequence to extract velocity informa- tion from the phase value of the moving spins. The most common phase technique is phase-contrast MRI. where Accurate quantification of flow with phase-contrast techniques ΔM is the perfusion-weighted magnetisation difference requires the flow encoding direction to be properly aligned with between the two experiments respect to the flow direction. Additionally, in the case of phase- α is the degree of inversion (for complete inversion α = 1) contrast MRI, the spatial resolution needs to be high in order to M t 0 is the tissue equilibrium magnetisation minimise the effects of intravoxel dephasing and partial-volume f is the tissue blood flow effects and to have a suitable number of voxels in the blood vessels. λ is the tissue-to-blood partition coefficient For more information, see Flow imaging and Inflow effect. T t 1 is the T1 of tissue Related Articles: Flow void, Inflow effect, Phase contrast Further Readings: Higgins, C. and A. de Roos. 2006. MRI and CT of the Cardiovascular System, 2nd edn., Lippincott Related Articles: Perfusion imaging, Arterial spin labelling, Williams & Wilkins, Philadelphia, PA; Tang, C., D. Blatter and EPISTAR, PICORE, QUIPSS – QUIPSS II – Q2TIPS D. Parker. 1993. Accuracy of phase-contrast flow measurements Further Readings: Kim, S. G. 1995. Quantification of relative in the presence of partial-volume effects. JMRI 3:377–385. cerebral blood flow change by flow-sensitive alternating inversion recovery (FAIR) technique: Application to functional mapping. Flow-sensitive alternating inversion recovery (FAIR) Magn. Reson. Med. 34:293–301; Kwong, K. K., D. A. Chesler, (Magnetic Resonance) FAIR is an arterial spin labelling (ASL) R. M. Weisskoff, K. M. Donahue, T. L. Davis, L. Østergaard, technique for non-invasive perfusion measurements, in which tis- T. A. Campbell and B. R. Rosen. 1995. MR perfusion stud- sue magnetisation is inverted in both a slice-selective (ss) and a ies with T1-weighted echo-planar imaging. Magn. Reson. Med. non-selective (ns) experiment, while the arterial magnetisation 34:878–887. is inverted only in the non-selective experiment (Figure F.39). In the slice-selective experiment, a 180° RF pulse is applied to the Flow void imaging slice, and after a labelling delay TI, the image is acquired (Magnetic Resonance) The term flow void is used for a frequently in the same region. In the non-selective experiment, the magne- observed signal loss phenomenon in MRI, caused by flowing tisation of the imaging slice as well as of the inflowing blood is material such as blood in vessels. Flow voids can have at least two inverted and the image is acquired (after the same delay TI as in technical explanations: the slice-selective experiment). The subtraction of the ns-based image from the ss-based image gives a perfusion-weighted dif- 1. In pulse sequences, where at least two RF pulses are ference map. The FAIR labelling technique is optimal in cases executed per repetition time interval (TR), a through- where the vascular geometry is complex, for example in the heart, plane motion of material during sequence execution lungs and kidneys. Perfusion quantification using FAIR is accom- will have the

effect that not all spins experience the plished by the following equation: desired pulse combination, thereby causing signal loss. For example, consider a spin-echo-sequence where ( ) = 2a t f t DM TI M0 TI × e-TI /T1 l blood flows in a vessel oriented perpendicular to the Fluence optimisation 382 Fluorescent screen imaging plane (figure below). The blood experiencing Fluorescence the 90° pulse (hatched) will, during the time period (Diagnostic Radiology) Fluorescence is the emission of a photon between the 90° and the 180° pulse (TE/2), move out of from a material in response to excitation by an incident photon of the slice and the spin echo is formed only in the part of higher energy. The wavelength of the emitted photon is character- the blood vessel that experienced both pulses (crossed); istic of the fluorescent material. thus, the signal is reduced in proportion to the amount A narrow light spectrum is emitted during fluorescence (with of blood that left the slice during TE/2. This phenom- very short afterglow ∼ ns). In medical physics, fluorescent materi- enon is also commonly known as the wash-out effect. als are used mainly for PM detectors and image intensifier input It should be noted, that in sequences requiring only screens (e.g. CsI:Tl). one RF pulse per TR, such as a conventional Gradient In contrast, broad light spectrum is emitted during phospho- Echo (GRE), the wash-out effect can normally be rescence (light continues after the radiation). In medical physics, ignored. phosphorescent materials are used mainly for monitor screens and image intensifier output screens (e.g. ZnCdS:Ag). 90° - TE/2 - 180° Materials exhibiting fluorescence in response to incident x-ray Imaging slice photons include zinc sulphide, calcium tungstate, barium lead sulphate and some rare earth compounds. The emitted wavelength from these materials is in the visible range. Fluorescent materials are employed in the screens used in film/screen combinations to convert incoming x-ray photons to visible light photons that in turn blacken the film. Blood vessel Fluorescent screen F (Diagnostic Radiology) In film screen x-ray imaging, the film is 2. When magnetic field gradients are applied, spins mov- in contact with an intensifying fluorescent screen. The purpose of ing in the gradient direction exhibit a phase evolution the screen is to convert incident x-rays to visible light that in turn that differs from that of static spins (see Velocity map- blackens the film (Figure F.40). The screens are made from mate- ping). After execution of a specific gradient pulse wave- rials which exhibit fluorescence, typically rare earth phosphors. form, the phase offset for spins in motion will depend on The phosphors have high atomic numbers and absorb x-rays much the specific gradient waveform as well as the velocity. more efficiently compared with film alone. Only a very small Hence, the phases will disperse in a volume of interest fraction of the latent image on the film is formed directly by the (voxel) covering a distribution of velocities, and the net x-ray beam – nearly 99% is due to light photons emitted from the signal will be reduced. The magnitude of this dephasing intensifying screen. effect may be substantial, and it depends upon imaging Screens are built into the light-tight cassette used to hold the parameters, such as voxel size and gradient strengths, as film. Usually, there is a screen on both sides of the film. On clos- well as upon motion direction and magnitude. ing the cassette, the screens come into close contact with the film surface. Use of screens on either side of the film yields better reso- Related Article: Velocity mapping lution than a single screen of the same total thickness. Where the efficiency of conversion of x-ray energy to light (the screen’s con- Fluence optimisation version efficiency) is very high, the upper screen may be thinner (Radiotherapy) Fluence optimisation is a technique used in radio- than the lower screen to avoid overexposure by the upper screen therapy treatment plan optimisation, particularly in external relative to the lower screen. beam x-ray planning. For each treatment beam, an optimisation The ‘speed’ of a film screen combination is a measure of its sen- algorithm is used to generate the fluence profile needed to deliver sitivity to x-ray exposure. Increased screen thickness, increased the prescribed dose distribution using inverse planning. After the phosphor size and increased phosphor efficiency increase the film plan has been optimised, a second stage is often involved in which screen speed. Resolution is predominately a function of the screen the constraints of the delivery system are modelled to produce a rather than the film, as the film has much higher resolution than deliverable beam plan. Related Articles: Interactive planning, Inverse radiotherapy planning, Simulated annealing algorithm Incident x-ray Fluid attenuated inversion recovery (FLAIR) photon (Magnetic Resonance) The aim of the FLAIR sequence is to suppress fluid signal by an inversion recovery preparation at an adjusted inversion time (TI). The TI value in the IR sequence Screen Light is chosen in a way that the signal of the fluid signal will have a emitted in zero crossing (nulling) due to T1 relaxation. The inversion time depends on the T1 relaxation time and is given by TInull ≈ 0.69T1 Film screen for sequences with long repetition time TR. Since the T1 relax- ation time of fluid is large, a long inversion time on the order of Screen 2000 ms is used in FLAIR. For further details, see STIR. Related Articles: Short tau inversion recovery (STIR), FIGURE F.40 The screen absorbs x-ray photons and emits photons in Inversion time (TI) the visible range that blacken the film. Fluorescent x-rays 383 Fluorine-18 (18F) the screen. A thicker screen, larger phosphor size and poor film to Further Reading: Curry, T., J. E. Dowdey and C. R. Murry. screen contact all reduce resolution. 1990. Christensen’s Physics of Diagnostic Radiology, 4th edn., Materials used as screen phosphors include rare earth phos- Lea & Febiger, New York. phors (e.g. lanthanum oxybromide, gadolinium oxysulphide) and calcium tungstate. Rare earth screens provide higher speed than Fluorescent yield calcium tungstate. Calcium tungstate and lanthanum screens emit (General) In an electronic transition from a higher shell to a in the blue region of the spectrum, while gadolinium phosphors vacancy in a lower shell, energy is released. The energy may be emit in the green. released as a characteristic photon (an x-ray if energetic enough) Nearly all x-ray interactions with the phosphor material are or an Auger electron. In Auger emission, the energy difference photoelectric. The number of x-ray photons absorbed increases between the two levels is imparted to an outer shell electron suddenly for x-ray energies above the K-edge of the phosphor. The which is ejected from the atom (Figure F.42). K-edge of a rare earth screen (39 keV for lanthanum, 50 keV for The fluorescent yield (ω) is the probability than a transition gadolinium) is lower than that of a calcium tungstate screen (70 results in emission of a photon. The probability that the transition keV). For all x-ray energies between the K-edges of rare earth results in emission of an Auger electron is 1 – ω. and calcium tungstate, the rare earth screen absorbs x-rays more efficiently. For the range of x-ray energies used in diagnostic radi- ology, rare earth screens are more efficient, absorbing 50%–60% Fluorine of the incident beam, compared with 20%–40% for a calcium (General) tungstate screen. Symbol F Fluorescent x-rays Element category Halogen (General) Fluorescent x-rays are x-rays emitted from a material Mass number A 19 in response to excitation of electrons in some manner. Fluorescent x-rays are characteristic of the material, and result from the same Atomic number Z 9 F process of electronic transitions that produce characteristic x-rays Atomic weight 18.998 g mol−1 in an x-ray tube. Usually, the term fluorescent implies that the Electronic configuration 1s2 2s2 2p5 excitation source is high-energy photons such as gamma rays or Melting point 53.53 K x-rays rather than a stream of electrons. Boiling point 85.03 K An incident high energy photon with sufficient energy can Density near room temperature 1.7 g L−1 (gas) eject an inner-shell electron. The vacancy resulting from ejection of the electron is filled by an electron falling from a higher shell. The excess energy between the two levels is emitted as a photon. History: Fluorine was first identified as part of the compound The energy gap between any two electron levels is characteristic fluorspar (calcium fluoride). This was initially described in 1530 of the atom. In tungsten, for example the primary transition from by Georgius Agricola in its capacity of catalyst in the compound- the L to K shell results in emission of an x-ray with an energy ing of various metals. It was also found to be useful in the process of 59.3 KeV (Curry et al., 1990). There is more than one energy of etching glass, when combined with acid. Due to its extremely associated with each electron shell, and transitions to a vacancy in high reactivity, the highest of any known element, it proved dif- an inner shell can take place from a number of shells. This results ficult to isolate in its elemental form, but this was finally achieved in a number of characteristic lines in the energy spectrum for a in 1886 by Henri Moissan, by electrolysis of hydrofluoric acid. particular atom (Figure F.41). Isotopes of Fluorine: A 100% of naturally occurring fluorine is found as stable 19F. It is an extremely reactive gas and is sel- dom found in its elemental form, which exists as F2, but readily Ejected forms compounds with many other elements. Radioactive 18F is electron the most common of the PET imaging tracers and is used for a variety of different studies. Medical Applications: PET imaging radioisotope – 18F decays by positron emission (97%) and electron capture (3%) to form 18O. Its high positron yield and a half-life of 110 min make Incident 18F highly suitable for physiological imaging studies using PET. photon The fluorine radionuclide is usually introduced into the body as K part of a larger molecule, such as fluorodeoxyglucose (FDG), and L 18F-FDG is the most widely used radiopharmaceutical in clinical PET studies. Related Articles: Positron emission tomography (PET), M Fluorodeoxyglucose (FDG), Fluorine-18 Characteristic Fluorine-18 (18F) x-ray (Nuclear Medicine) Element: fluorine FIGURE F.41 Ejection of an inner shell electron by an incident photon. • Isotopes: 15 < N < 24 Transition of a higher shell electron to fill the resulting vacancy results • Atomic number (Z): 9 in emission of a photon with energy equal to the energy gap between the • Neutron number (N): 9 two shells. • Symbol: 18F Fluorine (19F) imaging 384 Fluoro-glass dosimeter Auger electron K K L L M M Characteristic x-ray (a) (b) FIGURE F.42 (a) Electronic transition resulting in emission of a characteristic x-ray. (b) Electronic transition resulting in emission of an Auger electron. • Production: Cyclotron, e.g. 18 b+ O(p,n)18 F ® 18O or 1999. Table of Isotopes, 8th edn., Update with CD-ROM. http:// F + 20 e d,a 18 b 18 110min N ( ) F ® O ie .lbl .gov /toi .html]; Kowalsky, R. J. and S. W. Falen. 2004. 110min Radiopharmaceuticals in Nuclear Pharmacy and Nuclear • Daughter: 18O Medicine, 2nd edn., American Pharmacists Association, • Half-life: 109.77 min Washington, DC; Saha, G. B. 2004. Basics of PET Imaging: • Decay mode: β+ (96.9%), EC (3.1%) Physics, Chemistry, and Regulations, Springer, New York. • Radiation: β+ (max 640 keV), annihilation photons • Photon energy: 511 keV (194%) Fluorine (19F) imaging • Dose rate from 1 MBq: 120 μSv h−1 at 30 cm (point (Magnetic Resonance) Fluorine 19 (19F) has a nuclear spin of ½ source); 0.158 μSv h−1 at 1 cm (10 mL vial) so it experiences the nuclear magnetic resonance effect. 19F also • Absorption (range of β+): 0.6 mm in tissue has a relatively high gyromagnetic ratio (40.08 MHz/T), so it is • Critical organ: lungs (inhalation), stomach wall a good candidate for magnetic resonance imaging. However, the (ingestion) low natural abundance of fluorine makes it difficult to image in • ALImin (50 mSv): 2000 MBq the body. Therefore, the vast majority of 19F imaging is done with • Effective dose: 0.049 mSv MBq−1 (oral); 0.03 – 0.057 administration of an exogenous fluorinated contrast agent. The mSv MBq−1

(inhalation) 19F MRI contrast agents have application in cell targeting, cell in tracking, monitoring inflammation, and other applications. The development of fluorinated probes and polymer agents is an active 9 2P°3/2 area of research. A special ‘dual tuned’ receiver coil which enables imaging of F 1+ 109.77 m both 1H and 19F is often used to create ‘hot spot’ images where Fluorine 19F images overlaid on 1H images. 18 18 9 8 0 2 9F . 9 4 3 Further Reading: Jirak, D. et al. 2019. Fluorine polymer 1s2 2s2 2p5 EC probes for magnetic resonance imaging: Quo vadis? Magn. 17.4228 Reson. Mater. Phys. Biol. Med. 32:173–185. Fluorodeoxyglucose Clinical Applications: Fluorine-18 is often produced by irra- (Nuclear Medicine) 2-deoxy-2-[F-18]fluorodeoxyglucose (FDG) diation of 18O water using 10–18 MeV protons in a cyclotron. It is is a radiopharmaceutical used for PET imaging. FDG is a deriva- recovered as 18F sodium fluoride by passing the irradiated water tive of glucose, the predominant energy source for most cells of target through a carbonate type anion exchange resin column, and the body and for tumours. After intracellular phosphorylation via the 18F is retained on the column matrix. It is eluted with potas- hexokinase, FDG-6-phosphate is not significantly metabolised sium carbonate solution. and remains trapped in the cell. 18F sodium fluoride is used for the synthesis of 18F fluorode- oxyglucose (18FDG) as well as other 18F-labelled PET radiophar- Fluoro-glass dosimeter maceuticals. It is also used for bone scintigraphy, since it localises (Radiation protection) A fluoro-glass dosimeter is an accumu- in bone by exchanging with PO-4 ion in the hydroxyapatite crys- lation type solid state dosimeter based on the radiophotolumi- tal, resulting in PET images superior to 99Tcm MDP scintigraphs. nescent (RPL) phenomenon of silver-activated phosphate glass Further Readings: Chu, S. Y. F., L. P. Ekström and R. B. exposed to ionising radiation. Firestone. 1999. The lund/LBNL nuclear data search. http: / When silver activated phosphate glass is exposed to ionising /nuc leard ata .n uclea r .lu. se /nu clear data/ toi/; Firestone, R. B. radiation and later excited by ultraviolet ray, the glass emits an Fluoroptic® probe 385 Fluoroscopic portal imaging orange luminescence. The luminescence is proportional to the and FOV. The fluoroscopic dose rate is a parameter adopted to radiation dose that is exposed to the glass (Figure F.43). control the dose delivered to the patient, according to the selected When PO4 tetrahedron in glass is exposed to radiation, it mode. loses its electrons and traps positive holes (hPO4). The Ag+ ion in Depending on the technical publication adopted (IEC, FDA, glass traps a single electron and changes to the Ag0 ion. hPO4 is etc.), the dose rate reference points, the limits and the measuring assumed to move to the Ag+ ion and generates Ag++ ion, and these conditions are different. can function as the stable luminescence centre (Figure F.44). Fluoroscopic dose rate can be measured: Because the RPL effect is not erased by the reading opera- tion, repeatable reading can be performed unlimited times. The • At the entrance of the image receptor (without an anti- homogeneity of the glass reduces the difference among elements; scatter grid), using a copper filter to simulate the patient therefore, fluoro-glass dosimeters have excellent uniformity (tolerance level: 0,8 μGy s−1 for an image receptor with among glass elements. The stability of the RPL centre is high, a diameter of 25 cm; for alternate diameters, the limit is and there is little fading (it is not influenced by an environmental adapted according to the inverse square law) temperature). • At the entrance of appropriate phantoms (usually Perspex to simulate the patient), in order to predict Fluoroptic® probe entrance skin dose to the patient (with a 25 cm thick (General) ‘Fluoroptic’ is a registered trademark of LumaSense phantom, the maximum skin dose, including backscat- Technologies who make a range of fibre-optics-based temperature tering, shall not exceed 100 mGy min−1) sensing transducers and meters. • In air, 30 cm from the image receptor, using a lead fil- Fibre-optic-based thermometers can have the advantage that ter to reach the maximum dose rate (the limit is 10 R the temperature sensor element may be made electrically non- min−1) conductive and be unaffected by electromagnetic interference • 15 cm back from isocentre or high magnetic fields. They may, therefore, be suitable for F making measurements in harsh environments such as in MRI Fluoroscopic portal imaging scanners. (Radiotherapy) Portal imaging is the acquisition of images using Related Articles: Temperature control, Temperature probe a therapeutic x-ray beam in order to verify the beam delivery to Further Reading: http://www .lumasenseinc .com, Website the desired point of application during a radiotherapy treatment. accessed 31 July 2012. A common device commercially available is the fluoroscopic portal imaging: a metal plate is combined with a fluorescent Fluoroscopic dose rate screen to convert x-ray photons to light photons, which are viewed (Diagnostic Radiology) Fluoroscopic systems can offer different and digitised by a video camera. level of operation (modes) – continuous fluoroscopy, pulsed fluo- The metal plate is typically 1–2 mm thick copper or steel or roscopy at a range of frame rates, different x-ray beam filtrations brass; for the fluorescent screen, a rare earth phosphor (usually FIGURE F.43 Radiophotoluminescent phenomenon of silver-activated phosphate glass exposed to ionising radiation. FIGURE F.44 Generation of the RPL centre. (Image courtesy of Chiyoda Technol.) Fluoroscopic timer 386 F lying focal spot gadolinium oxysulphide) is generally adopted. A series of mirrors Fluoroscopy, digital reflects the light photons into a lens and camera, which are sur- (Diagnostic Radiology) See Digital fluoroscopy rounded by lead shielding. The device enables imaging at any gantry angle and can be Fluoroscopy, mobile attached/detached or retracted from the position of use. (Diagnostic Radiology) See Fluoroscopy The images are digital and can undergo image processing, image matching and digital archiving. Moreover, the images are Flux immediately available and can be used interactively to optimise (General) Flux is a measure of the rate of flow of some quantity field position during the treatment. through a defined surface area (Figure F.45). A recent developing application of electronic portal imaging Where the rate of flow of the quantity in question is defined by device (EPID) is their use for pretreatment dosimetric verifica- some vector field (V), flux can be defined as tion of intensity-modulated beams and determining patient dose information. Flux =òV × dS Fluoroscopic timer S (Diagnostic Radiology) Fluoroscopic equipment is often used to The formula represents integration of the component of the vector perform very complex and time-consuming diagnostic and inter- field aligned with the normal to each infinitesimal area (dS) over ventional procedures, which require special radiation protection the entire area of the surface (S). measures. Fluoroscopic x-ray systems are always provided with a timer. It is a cumulative timing device, switch activated by the Flux density fluoroscopic exposure. The fluoroscopic timer is used to prevent (General) Flux density is a measure of the rate of flow of some overexposure of the patient and to record the total length of a fluo- quantity through a unit area of a defined surface (Figure F.45). F roscopic procedure. Where the rate of flow of the quantity in question is defined by The irradiation is automatically terminated when the total some vector field (V), flux density can be defined as exposure time elapses – a predetermined limit, which should not exceed 10 min. Furthermore, an acoustic signal is activated after the integrated time has reached a limit not exceeding 5 min and at ò V × dS least 30 s before the automatic interruption of the exposure. Flux density = S A manual reset of the fluoroscopic timer is required both to ò dS interrupt the signal and to restart the exposure. S The accuracy of the fluoroscopic timer is verified during the The formula represents flux divided by the total area of S. quality control tests. Flying focal spot Fluoroscopy (Diagnostic Radiology) In CT, a flying focal spot is used to (Diagnostic Radiology) Fluoroscopy is a mode of use of x-rays increase sampling frequency and improve image spatial reso- to visualise moving organs or to perform dynamic examination. lution. It is sometimes also referred to as a dynamic focal spot This procedure requires a special x-ray detector. The earliest (Figure F.46). fluoroscopic equipment used just a fluorescent screen (phosphor) Conventionally, on a third-generation scanner, the number of as a detector. This gives off a faint light only visible in a dark samples per projection equals the number of detector elements in room. Later designs processed the image from the phosphor the x–y plane. Some CT scanners employ a flying focal spot to by a system of mirrors and lenses leading to 4 times minifica- double the number of samples per projection by sampling each tion and 16 times intensification of the image brightness. This detector element twice in every projection. To achieve this, the image was then passed to a superorticon TV camera and viewed position of the focal spot on the anode is electromagnetically on a TV monitor. Contemporary fluoroscopic systems use an deflected (by deflecting the beam of thermal electrons from the image intensifier (II), which allows a total amplification (gain) cathode, so these hit a different spot on the anode). This deflection of the imaging system of up to 7000. The II is again linked to of the focal spot is made after the gantry has rotated a distance a TV camera and monitor. It is expected that in future the flat panel detectors with thin film transistors technology (TFT) will replace the II. The image contrast in fluoroscopy depends on the x-ray spec- trum exactly in the same way as in radiography. Due to this reason, various organs are filmed (static image) or observed in motion (fluoroscopy) using similar kV ranges. A major advantage S of II is the greatly reduced intensity of the x-ray beam (normally an II system uses 0.3–3 mA anode current). However, the overall patient dose delivered during fluoroscopy is much greater than in radiography. This is due to the long examination times (several minutes). Observation of hollow organs in fluoroscopy requires the use of various contrast media such as barium meal for imaging of the gastrointestinal system, or iodine contrast for imaging the blood vessels and the heart (angiography). FIGURE F.45 Flux is the rate of flow of some quantity through a defined Related Articles: Image intensifier, Cinefluoroscopy surface area. fMRI (functional magnetic resonance imaging) 387 Focal point Anode Linear tube motion α = tomograpphic angle α Fulcrum or focal plane FIGURE F.46 Showing the original projection measurement (on left, in darker grey), and with focal spot deflection after gantry rotation equal to half detector element (on right), where the original is still in darker grey, while the half detector position is in lighter grey. (Courtesy of ImPACT, UK, www .impactscan .org) Linear receptor motion equal to half a detector element; after this, the focal spot returns FIGURE F.47 Focal plane tomography principle. to the position it was at for the previous projection measurement. This results in two interlaced measurements per projection. The flying focal spot may also be available in the scan axis The level of the fulcrum can be selected according to the (z-axis) direction, such that two overlapping slices are acquired in height of the anatomy of interest. each tube rotation, effectively doubling the sampling rate to half a Furthermore, the section thickness is inversely proportional to F detector width in that dimension. the tomographic angle. The thinner is the slice, the more detail for small region will be detectable. fMRI (functional magnetic resonance imaging) The focal plane tomography was introduced in 1921 and was (Magnetic Resonance) See Functional magnetic resonance imag- used intensively until the 1980s. ing (fMRI) More common diagnostic procedures were intravenous pyelo- grams, IVP (used to visualise abnormalities of the urinary sys- Focal film distance (FFD) tem, including the kidneys, ureters, and bladder), temporal bone (Diagnostic Radiology) FFD is the distance between the focal imaging and chest radiographic imaging. spot of the x-ray tube, or the radiation source, and the film, or the Now, focal plane tomography is almost completely replaced by image receptor. It is also known as TFD or source image receptor the computed tomography. distance (SID). Related Article: Classical linear tomography FFD is

an important parameter influencing the intensity of the x-ray beam – the smaller the FFD, the greater the intensity of the Focal plane tomography x-ray beam. (Nuclear Medicine) Focal plane tomography was used before This dependence is defined by the inverse square law – inten- emission computed tomography became widely available. sity of radiation decreases or increases by the square of the dis- Images were obtained at different projection angles and then tance from the source. shifted and superimposed so that one ‘longitudinal’ plane (par- allel to the camera face) was focused whereas all other planes were blurred. EXAMPLE One example of focal plane tomography used a collimator with multiple pinholes. The different pinhole images were shifted and Typical FFD values 80–110 cm superimposed to bring certain planes into focus. Another example Doubling the distance from the x-ray tube to the patient used a rotating slant-hole collimator. (D) will cover four times the area at 2D but reduces the The use of these techniques has mostly died out because x-ray intensity to 1/4th {(1/2)2} underlying and overlying blurred planes obscure the focused plane of interest. See the diagram in the article Target film distance (TFD). Related Article: Multiple pinhole Focal plane tomography Focal point (Diagnostic Radiology) Focal plane tomography is another name (Ultrasound) For imaging purposes, a narrow ultrasound beam for classical linear tomography. During the exposure, the tube is desirable for resolution. The directivity of an ultrasonic beam and the image receptor are in synchronous movement, while the can be altered by focusing. Although the general principles are patient remains still (Figure F.47). the same as applied in optics, one must be aware of issues related The resulting image shows the anatomical structures lying on to the wavelength and the situations when the dimensions of the the focal plane satisfactorily recorded, while the details above and transducer are comparable to the ultrasonic wavelength, as scat- below the focal plane appear blurred or eliminated. tering will occur. In this case, the laws of reflection and refrac- The reason is that during the exposure in the focal plane the tion still apply, but one might use the principle of superposition radiographic motion is minimum, and the region appears clearly and finite elements to model the resulting US field. Focusing can defined on the image. be performed with curved transducers, focusing lenses or mirrors Focal spot 388 Focal spot, effective Electron beam a Heat Focal sp ot tra ck h R Rotation Projected An focal spot a o n d g e FIGURE F.48 Focusing with single element curved transducers. le for a single element source or, as done in modern systems, with FIGURE F.49 Small section of a rotating anode with the different focal transducer arrays and electronic control of firing, delaying the spots. firing of certain array elements in relation to others to emulate the effect of a curved transducer. The obvious advantage of the electronically controlled array is that the focal distance can be relationship, large heat capacity and small projected size but limit varied. The distance from the source to the focal point (or focus) F the area that is covered by the x-ray beam (FOV) because of the is the focal length (F). Wells (1977) points out that in the case of a heel effect. Figure F.49 shows a small section of a rotating anode spherical concave radiator, vibrating with uniform normal veloc- with the different perspectives of a focal spot. ity and with a radius of the circular boundary, which is large in Related Articles: Anode, Cathode, Target, Line focus prin- relation to both the wavelength λ and the depth of the transducer ciple, Stationary anode, Focal spot actual, Focal spot effective (Figure F.48), the ratio of the intensity of ultrasound at the centre of curvature to the intensity at the radiating surface is given by Focal spot, actual (Diagnostic Radiology) The actual focal spot of an x-ray tube is I r ( 2 = kh) the area of the target bombarded by the beam of thermal elec- I0 trons. This area takes all thermal energy of the electron beam where k (the wave number) = 2π/λ and that the point of greatest and this way determines the power of the x-ray tube (the larger intensity is on the central axis but is not at the centre of curvature the actual focus, the more powerful is the x-ray tube). In classical R, as one would expect. The focusing can be improved by suitable x-ray tubes, this area is a projection of the shape of the filament choice of dimensions (Wells 1997). wire (coil). This way, it has approximately rectangular shape, Further Reading: Wells, P. N. T. 1977. Biomedical where the width is roughly equal to the diameter of the filament Ultrasonics, Academic Press, London, UK. coil (approximately 0.4 mm) and the height is determined by the length of the filament coil and the cosine of the anode angle. In Focal spot some contemporary x-ray tubes (with an electrical system which (Diagnostic Radiology) A focal spot is the area on the surface of focuses the thermal electrons), the actual focal spot can vary in an x-ray tube anode (target) that is bombarded by the high-energy shape and size. electrons from the cathode and where the x-radiation is produced. Some sources distinguish two parts of the actual focal spot – Two focal spots can be defined – actual focal spot and effective electronic focal spot (the section of the electron beam bombarding focal spot (see the eponymous articles). the target) and thermal focal spot (the area of the target which has The actual focal spot is the area of the target bombarded by the been bombarded – i.e. the actual focal spot). accelerated thermal electrons. This area takes all thermal energy In x-ray tubes with a rotating anode, the heat is distributed of the electron beam and this way determines the power of the over a ring of the anode target, called thermal track (Figure F.50). x-ray tube. In classical x-ray tubes, this area is a projection of the In this case, the actual focal spot is often called ‘real’ (or ‘momen- shape of the filament coil (i.e. approximately rectangular shape). tary’) focal spot (see article on Rotating anode). The thermal In some contemporary x-ray tubes (with electrical system which track distributes the temperature of the actual focal spot over a focuses the thermal electrons. e.g. Straton tube), the actual focal large area, thus effectively enlarging it (without enlarging the spot can vary in shape and size. effective focal spot) and allowing an increase of power of the The effective focal spot is the projection of the actual focal x-ray exposure. spot down the central x-ray beam (towards the patient). Its size Related Articles: Anode, Anode angle, Cathode, Target, depends on the anode angle (see Line focus principle) and deter- Stationary anode, Focal spot effective mines the resolution of the x-ray image. From optical point of view, the effective focal spot can be seen as the ‘x-ray point Focal spot, apparent source’. (Diagnostic Radiology) See Focal spot, effective The relationship between the actual focal spot on the sur- face of the anode and the projected effective focal spot size is Focal spot, effective determined by the anode angle. The angle is a design character- (Diagnostic Radiology) The effective focal spot is the projection istic generally in the range of 6°–24°. Small angles give the best of the actual focal spot down the central x-ray beam (towards the Focal spot, effective 389 F ocal spot, effective Anode disk W-Re coating Thermal path (LF+SF) Thermal path (LF) Mo anode stem α-Anode angle α Actual focal spot LF SF Graphite layer Cathode FF W-Re target coating F BF Mo anode body Effective focal spot Central x-ray beam towards the patient FIGURE F.50 Actual focal spots of an x-ray tube. Note the fact that if both cathode filament wires (for large focus [LF] and for small focus [SF]) overlap, then part of the actual focal spot may be overloaded. The effective focal spots (FF – fine focus and BF – broad focus) are towards the patient. patient) – Figure F.50. The effective focal spot determines the resolution of the x-ray image. From optical point of view, it can be seen as the ‘x-ray source size’. Some sources name the effective focal spot as apparent focal spot or as optical focal spot. Ordinary x-ray tubes (for general radiography or fluoroscopy) have almost square effective focal spot with size on the order of 0.6–1.2 mm (a larger focal spot will produce too blurred images). Mammographic x-ray tubes (producing high resolution images) have similar effective focal spot with size on the order of 0.2–0.4 mm. Some contemporary x-ray tubes use electrical system which focuses the thermal electrons, thus producing variable effective focal spot shape and size. The size of the effective focal spot Fe depends on the actual focal spot size (Fa) and the anode angle (α) (see Line focus principle): Fe = sina × Fa Obviously, a small anode angle will produce small size of the effective focal spot. However, this will reduce the size of the use- ful x-ray beam (the section of the beam). FIGURE F.51 Radiograph of the effective focal spot taken with pinhole Because the effective focal spot is a projection down the camera. The size of the image is often 1:1 the size of the eff. focal spot. central x-ray beam, its size is measured at the centre of the Normally, quite heavy exposure is required for this measurement. x-ray beam. Various tools are used for this purpose. The pin- hole camera is used for direct measurement – Figure F.51. Other special test objects (e.g. star test or bar pattern) use high resolu- spot size from the image of the test objects with high resolution tion patterns for indirect measurement – Figures F.53. All focal patterns. spot measurements require exact distances focus/test-object/ Figure F.52 shows a bar pattern test object used for indirect film (specific for the test object or pinhole camera). Also, spe- measurements of the effective focal spot size. In this case, the cific formulas are used for the calculation of the effective focal size of the focal spot is derived (through a table) from the size Focal spot selection 390 Focal spot selection FIGURE F.54 Real image of the effective focal spot of a new x-ray tube made with multiple pinhole camera. (Image courtesy of K. Kepler.) FIGURE F.52 Bar-pattern test object used for indirect measurements of the effective focal spot size. (Image courtesy of S. Tabakov.) F FIGURE F.55 Real image of the effective focal spot of an old x-ray tube made with multiple pinhole camera. (Image courtesy of K. Kepler.) from the x-ray tube). Aiming at better understanding, the central spot of the effective focal spot size on Figure F.53 has been shown as rectangular with significantly different sides. Figures F.54 and F.55 show real images of measurement of the effective focal spot made with multiple pinhole camera. On both images, the vertical direction is the cathode-anode direction. In Figure F.54, the measurements are from a new modern x-ray tube (production 2007); in Figure F.55, the measurements are from an old x-ray tube (production 1986). Note that the new x-ray tube presents overlapping images of both the fine focus (the smaller FIGURE F.53 Variation of the size and shape of the effective focal spot darker spot) and the broad focus (the larger paler spot). Both foci of an x-ray tube over the x-rayed field. of the old x-ray tube do not overlap (the fine is below the broad focus). The image of the new x-ray tube shows better the enlarge- ment of the effective focal spot. of the blurred patterns. Note that here two different patterns are Related Articles: Anode, Anode angle, Anode heel effect, blurred (the 8th and 9th patterns) showing that the focal spot is Cathode, Line focus principle, Stationary anode, Focal spot not square. actual, Pinhole camera In a large x-ray field, the effective focal spot size will increase its size towards the periphery of the field (away from the centre). Focal spot selection This phenomenon creates varying spatial resolution of the images (Diagnostic Radiology) The focal spot selection is made through (the best one being at the centre

of the x-ray beam). This variation the filament circuit of the high voltage generator. is most prominent in the cathode-anode axis of the tube, where Each focal spot is associated with a different filament wire. the Heel effect has an additional influence to the effective focal The selection takes into consideration the maximum permissible spot size and x-ray beam intensity – Figure F.53. Note – the figure temperature of the filament of the respective focal spot. presents only a rough idea about the effective focal spot size and The current through the cathode filament (filament current shape at the edges of the maximal x-ray field (at least 1 m away If) depends also on the maximum permissible power for selected Focal spot selector 391 Focused grid focal spot (Pmax). The filament circuit assures anode current In Figure F.56, only the part of the received data that is close Ia < Pmax/Ua for each selection of focal spot and anode voltage. to respective focus is used for image data. Note how a larger aper- The selection of a focal spot size for a specific clinical pro- ture is used for the deeper zones. cedure generally represents a compromise between two require- ments. Small sizes are desirable to minimise image blurring and Focused collimator provide good visibility of anatomical detail and small signs of (Nuclear Medicine) See Diverging collimator pathology. However, large focal spot areas are required to reduce heat concentration and permit the necessary x-ray exposure to be Focused grid produced for the procedure. (Diagnostic Radiology) A focused grid is designed so that the Selection of focal spot size occurs in two stages. The first is the spaces between the attenuating strips are at an angle so that they selection of the tube design to be used for the various clinical pro- align with or focus a point in space as shown in Figure F.57. cedures. Tubes with small focal spots are selected for procedures The focal distance (such as 1 m) is on the grid label along with that require high visibility of detail such as mammography. Tubes which is the x-ray tube side of the grid. with larger focal spots are selected for procedures that require the If the grid is not positioned correctly as shown in Figure F.58, production of relatively high exposures to penetrate and image the primary x-ray beam will not be aligned all over the grid and large body sections such as the abdomen and the lateral chest. the condition of ‘cutoff’ will occur. Most tubes are designed with two focal spot sizes. The opera- tor or the automatic exposure control function of the equipment select between the ‘small’ and the ‘large’ focal spot size for a spe- Grid focus cific procedure. Generally, the smaller size is used for good image detail if it has the necessary heat capacity. Grid focal point Related Articles: Cathode, Filament, Filament circuit, High voltage generator, Target, Focal spot actual, Focal spot effective x-ray tube focal spot F Focal spot selector (Diagnostic Radiology) See Focal spot selection Focal zone Focal distance (Ultrasound) In transmit mode, a number of separate transmis- sions must be used if lateral resolution is to be maintained over the entire imaging depth. For each transmission, the focus is placed at different depths, and only the part of the received data that is close to respective focus (referred to as the focal zone) is used for image data. In Figure F.56, three transmissions are used to make up the image. Note how the aperture is increased with increasing imag- ing depth to maintain the lateral resolution (F-number). FIGURE F.57 The positioning of a focused grid so that the grid focal point is located at the x-ray tube focal spot. (Courtesy of Sprawls Foundation, www .sprawls .org) X-ray tube Zone 1 focal spot Grid focal point Zone 2 Zone 3 FIGURE F.58 Two conditions that produce grid cutoff because the grid FIGURE F.56 Three different transmissions are used to achieve focus- focal point is not located at the x-ray tube focal spot. (Courtesy of Sprawls ing at three different positions. Foundation, www .sprawls .org) Focusing cup 392 F orce balance Focusing cup conduction bands (normally empty of electrons) are separated, (Diagnostic Radiology) This is a small metal cup which holds the trapping the electrons to their individual atoms and preventing cathode wire of the x-ray tube. Its function is to focus the thermal them from easily moving from atom to atom. electrons emitted by the cathode filament, which otherwise pro- In semiconductors, the energy needed to cross the forbidden duces quite spread beam of electrons towards the anode. This way, gap is only low, and this can be overcome with low electric poten- the focusing cup minimises the focal spot size over the anode. tials, leading to the use of semiconductors in electronic devices. The focusing cup is specially shaped and is made of molybdenum, Insulators have wide forbidden bands, which can only be over- nickel or steel, because of their poor thermionic emission. Placing come with strong electric fields, leading to common electrical the filament wire inner or outer in the cup (during the produc- insulators. tion of the x-ray tube) changes the focusing of the electron beam. Common electrical conductors such as copper, silver and gold When negative voltage is applied to the focusing cup, it becomes have no forbidden band, and their electrons can freely move in a Wehnelt electrode (see the eponymous article). Changing the and out of the valence band, providing easy electrical current flow voltage of this electrode changes the shape of the beam of thermal with very little electric field needed. electrons, hence the focal spot size. This electrode can also stop It is interesting to note that while conductors have a resistance the beam of electrons (in the grid-controlled x-ray tubes). to current flow which rises slightly with temperature, one of the Related Articles: Cathode, Wehnelt electrode, Focal spot, defining properties of semiconductors is that their electrical resis- Grid-controlled tube tance decreases significantly with increasing temperature. Foetus Force balance (General) In humans, the foetus is the unborn young from the (Ultrasound) Force balances are used to measure the power end of the 8th week after conception to the moment of birth, as output of ultrasound devices. The total acoustic power passing distinguished from the earlier embryo. In a foetus, all major body through an area is defined by the integration of the intensity over F organs are present. that area. This can be expressed as a summation: See also Embryo W = å Ita × DA Fog (Diagnostic Radiology) See Film fog where Ita is the temporal average intensity Foot switch ΔA is the square of the sample distance (Diagnostic Radiology) See Dead man’s switch The summation is performed for all Ita (>1% of the peak) mea- Forbidden energy gap sured in the specified plane. This parameter can be measured with (General) All atoms normally have electrons surrounding them; a hydrophone system but this is complex and time-consuming. however, there are only a fixed number of possible stable orbits or The method usually preferred is to use a force balance. shells where they can exist. Electrons may move from inner orbits The ultrasound beam exerts a force on the balance. There to outer orbits or be ejected completely if they are given enough are two basic designs of force measurement device, using either energy through absorbing electromagnetic radiation. Low orbits absorption or reflection of the ultrasound beam (Figures F.59 and require more energy to escape, whilst outer orbits need less. F.60). The limited number of orbits means that the electrons around For a plane wave incident on a perfectly absorbing target, the atoms have a clearly defined ability to absorb/re-emit radiation radiation force is given by only at certain fixed energies equalling the energy difference between allowed orbits. Above this layer of valence electron IA W orbits lie further unfilled orbits to which excited electrons may F = = c c be energised to fill. This defines the typical bands seen in optical absorption/emission spectra of elements in gaseous form. In solids, the nearness and binding forces between atoms and the possible mix of elements result in these individual energy lev- els being reduced to allowable energy bands: Conduction band Conduction Conduction band band Forbidden band Forbidden band Energy Valence Valence Valence a F F ban band b nd d Absorber force balance Reflector force balance Isolated atoms Insulators Semi-conductors Conductors Allowable electron energy levels and bands in carious materials The forbidden energy bands of some solids means that in some FIGURE F.59 Absorbing (L) and reflecting (R) force balances. (Courtesy cases the valence bands (normally filled with electrons) and the of EMIT project, www .emerald2 .eu) Force, electrostatic 393 F our-rectifier circuit Absorbing target 2.250 FIGURE F.61 Water bath supported on the table separately from the target which is supported by the weighing device. This allows use of more appropriate scale of weighing device. (Courtesy of EMIT project, www .emerald2 .eu) = 8.9875517873681764 ´109 Nm2 C-2 F FIGURE F.60 Force balance in operation. (Courtesy of EMIT project, www .emerald2 .eu) where c0 is the speed of light in a vacuum equal to 299,792,458 m s−1 μ0 is the magnetic constant defined as 4π × 10−7 H m−1 where c is the speed of ultrasound in the propagation medium, ɛ0 is the electric constant equal to 1/(m c2 0 0 ) ∼ 8.854187817 × −12 usually water. 10 F m−1 It may be difficult to achieve perfect absorption, and an alter- native method is to use a conical air-filled reflector. For a perfect Related Articles: Charge, Coulomb reflector, the force and power are related by Further Reading: Resnick, R. and D. Halliday. Physics, Wiley International Edition, New York. c × F W = 2cos2q Forces, nuclear (General) See Nuclear forces where θ is the angle between incident beam and the normal to the reflecting surface. Forensic radiology The forces that result from diagnostic and therapeutic devices (Diagnostic Radiology) Forensic radiology and post-mortem are small, typically mN. Sensitive weighing devices are used to imaging are methods used in criminology. These use most of the measure the slight change in weight. To improve sensitivity, the diagnostic medical imaging equipment (x-ray, MRI, ultrasound, target may be suspended on a separate structure linked to the bal- isotope imaging, etc.) and give additional information to the crim- ance to remove the weight of the water and container (Figure F.61). inal investigation. Related Articles: Acoustic power, Acoustic pressure, Absorption, Intensity Forward treatment planning (Radiotherapy) Forward planning is a conventional radiotherapy Force, electrostatic treatment planning technique in which the planner selects by expe- (General) The magnitude of the electrostatic force (F) on a charge rience the required number of open and/or wedged treatment beams (q of appropriate beam geometries. He then calculates either manu- 1) due to the presence of another charge (q2) is given by ally or by means of a treatment planning computer the compos- ite distribution of dose by adding the dose contributed by each of q q F = k 1 2 e the treatment beams. If the dose and the distribution of dose in the r2 irradiated volume are unsatisfactory, the planner varies the beam where parameters and geometries and repeats the calculation. The pro- r is the distance between the two electric charges q1 and q cesses are repeated until an acceptable treatment plan is achieved. 2 ke is proportionality constant Four-rectifier circuit (Diagnostic Radiology) Four-rectifier circuits, also known as The proportionality constant ke is called Coulomb’s constant bridge rectifier (or Graetz circuit), consist of four diodes con- and is given by nected in a way to form a bridge. This arrangement allows full wave rectification by converting the input AC voltage into a posi- 1 m c2 tive pulsating DC voltage (Figure F.62). ke = = 0 0 4pe0 4p See Rectifier Fourier reconstruction 394 F ractionation different frequencies. The Fourier transform can also refer to the function’s representation in the frequency domain. To read more about Fourier analysis and Fourier series, follow the external links. Hyperlinks: http: / /mat hworl d .wol fram. com /F ourie rSeri es .ht ml; http: / /mat hworl d .wol fram. com /F ourie rTran sform .html FOV (field of view)

(Nuclear Medicine) See Field of view (FOV) FIGURE F.62 Four-rectifier circuit. Fractional anisotropy (FA) (Magnetic Resonance) FA is a scalar index for diffusion Fourier reconstruction anisotropy. It is determined from the eigenvalues of the diffusion (General) See Filtered back projection tensor and is defined as Fourier spectrum framework æ 2 2 2 3ç(l - l) + ( - ) + ( - ) ö÷ é 2 1 l2 l l3 l 3( i ù = è ø ëål = û - l2 (Magnetic Resonance) In MRI, there is a direct relationship ) FA between the collected data and the reconstructed image through 2(l2 + 2 + 2 1 l2 l3 ) 2ål2 i Fourier theory and the Fourier transform. The same relation- ship is used in nuclear magnetic resonance (NMR) spectroscopy where which is taken as the example here since it is limited to one λi is the ith eigenvalue of the diffusion tensor F dimension. l is the mean of the eigenvalues, i.e. the mean diffusivity Briefly, signal from a specified volume of tissue is acquired (MD) after the transmission of an RF pulse (excitation). This yields the FID, i.e. signal as a function of time. After applying the The FA ranges from 0 (isotropic diffusion) to 1 (anisotropic Fourier transform to the detected signal, a spectrum revealing diffusion). The FA is a rotationally invariant metric, i.e. it is inde- its frequency content is obtained. Mathematically, this can be pendent of the major diffusion direction and the orientation of the considered a decomposition of the initial signal (time domain) object (Figure F.63). into an infinite sum of sines and cosines with different frequen- Related Articles: Diffusion tensor, Relative anisotropy cies (frequency domain). The signal as a function of time and the frequency spectrum is a Fourier pair and the signal can Fractionation be written as the inverse Fourier transform of the frequency (Radiotherapy) Generally, radiotherapy treatment is delivered spectrum: once a day, 5 days a week for up to 8 weeks. Each daily treat- ment is called a fraction and such regimes have been shown to ¥ f (t ) = òF (u)ei2put be clinically effective and acceptable, giving a favourable thera- du (F.3) peutic effect in most cases: better tumour control is obtained -¥ for a given level of normal tissue toxicity when the radiation where dose is fractionated rather than delivered as a single dose. The f(t) is the FID u is the frequency variable F(u) is the complex data represented in the frequency domain This can also be written as F (u) = F (u) eiq(u) (F.4) The magnitude F(u) is called the Fourier spectrum and θ(u) is the phase angle. In spectroscopy, a spectrum allows visualisation of the rela- tive occurrence of various metabolites, which, due to their vary- ing chemical environment, contain protons with correspondingly varying resonance frequencies. Related Articles: Fourier transform, Free induction decay, Frequency spectrum Four-dimensional (4D) dose calculation (Radiotherapy) See 4D dose calculation Fourier transform (General) Fourier transform refers to the process of decompos- FIGURE F.63 A fractional anisotropy map obtained from a healthy ing a function in terms of the sum of sinusoidal functions with volunteer. Fractions 395 Frame mode effectiveness of fractionated radiotherapy can be understood by Abbreviations: BED = Biological effective dose and EQD2 = consideration of the 5Rs of radiobiology. Dividing the radio- Equivalent total dose in 2 Gy fractions. therapy dose into fractions spares normal tissues since the cells Related Articles: Alpha beta ratio, Biological effective dose, can repair some of the radiation damage in the time between Cell cycle, Fractionation, Interruption of treatment, Linear fractions, and cell repopulation will occur provided the over- quadratic (LQ) model, Radiosensitivity, Repair, Repopulation, all time is sufficiently long. Conversely, fractionating the dose Reoxygenation, Therapeutic effect, Tolerance, 5Rs of increases the damage to the tumour due to reoxygenation and radiobiology the redistribution of cells into radiosensitive phases of the cell Further Readings: Bentzen, S. M. et al. 2008. The UK stan- cycle between fractions. dardisation of breast radiotherapy (START) trial B of radiotherapy Many of the regimens currently in use have developed as a hypofractionation for treatment of early breast cancer: A ran- result of expediency rather than from radiobiological principles. domised trial. Lancet 29:1098–1107; Hall, E. J. and A. J. Giaccia. Historically, the most common dose per fraction in use has 2006. Radiobiology for the Radiologist, 6th edn., Lippincott been 2 Gy. However, in the Northern United Kingdom, frac- Williams & Wilkins, Philadelphia, PA; Nias, A. H. W. 1998. An tion sizes of 2.67–2.75 Gy per day have been used for many Introduction to Radiobiology, 2nd edn., John Wiley & Sons Ltd, years, introduced largely to ease the burden on thinly spread Chichester, UK; Saunders, M. et al. 1999. Continuous, hyperfrac- resources. It should be noted that the total dose delivered with tionated, accelerated radiotherapy (CHART) versus conventional these hypofractionated regimens is reduced and that the clinical radiotherapy in non-small cell lung cancer: Mature data from results have generally been shown to be comparable to those for the randomised multicentre trial. Radiother. Oncol. 52:137–148; the more standard 2 Gy per fraction protocol. Indeed, there is The Royal College of Radiologists. 2006. Radiotherapy Dose- renewed interest in hypofractionation in the light of new data Fractionation, London, UK. on the alpha-beta ratio values for some tumours, the results of fractionation trials such as START and the advent of intensity- Fractions modulated radiation therapy. More details can be found in the (Radiotherapy) Generally, radiotherapy treatment is delivered article on Hypofractionation. F once a day, 5 days a week for up to 8 weeks. Each daily treat- As discussed earlier, the prolongation of treatment by the use ment is called a fraction. For more information on the rationale of fractionation has been shown to be beneficial in many cases. of fractionation, see the articles on Fractionation and The 5Rs of However, the excessive prolongation of treatment may actu- radiobiology. ally have a detrimental effect on the therapeutic efficacy of the Related Articles: Fractionation, 5Rs of radiobiology treatment if proliferation of tumour cells becomes significant. Irradiation of tumour cells can trigger the surviving cells to divide faster than before. This is known as accelerated repopula- Frame mode tion (further details can be found in the article on Repopulation). (Nuclear Medicine) One of the two modes of acquiring images in Hyperfractionation, particularly accelerated hyperfractionation, nuclear medicine (the other being list mode). Before image acqui- offers a possible solution to this problem. Continuous hyperfrac- sition, a portion of the image acquisition computer’s memory is tionated accelerated radiation therapy (CHART), which reduces reserved to contain the image, and all pixels within this image are overall treatment from 6–7 weeks to 12 days and gives 36 small set to zero. As the X and Y position signals of each scintillation fractions, has been tested in multicentre randomised controlled event are received from the gamma camera, a count is added to clinical trials. The trial in non-small-cell lung cancer showed the equivalent pixel in the computer memory. Counts are added as improvement in survival, and this regimen is now the govern- the subsequent X and Y position signals, and an image is formed, ment-recommended standard of care for eligible patients in the which is then sent to the video monitor for display via a video United Kingdom. More details can be found in the article on interface. Hyperfractionation. There are three types of frame mode acquisitions: In the United Kingdom, the Royal College of Radiologists published a report in 2006 in which they identified fraction- 1. Static mode – A single image is acquired for a preset ation regimens for which there is high-quality evidence for both time or until preset number of counts are reached. safety and efficacy. They found a state of equipoise in many 2. Dynamic mode – A series of images are acquired in clinical situations, meaning that there is genuine uncertainty series, with each image being acquired for a fixed over whether or not the treatment will be beneficial, where pub- amount of time per image. lished evidence was insufficient to favour one particular regi- 3. Gated mode – This mode is used when there is a men over another and affirmed that clinical trials are required dynamic, repetitive process such as a beating heart. A to resolve these issues. physiological trigger such as the patient pulse from the In cases where a change in fractionation regimen is consid- ECG monitor signals the start of acquiring a set number ered, perhaps due to an interruption of treatment or as part of a of images, which are stored in separate time bins in the clinical trial, it is useful to be able to compare the regimens in computer memory. The system then waits till it receives terms of the effect on both the tumour and the normal tissues. the next trigger, and the process is repeated, and each This is possible by using the linear-quadratic model to evaluate of the time bins is filled with images of the organ being the biological effective dose (BED). In the case of normal tissues, in a specific position. After acquiring for a preset time, a more useful parameter may be the equivalent total dose in 2 Gy the images can be played in series like a movie, which fractions (EQD2) since most clinical experience of normal tis- mimics the physiological repetitive process that is being sue tolerance has been obtained from 2 Gy per fraction regimens. studied. Gated mode imaging is most commonly used More detail can be found in the article on BED. for cardiac studies. Frame mode for digital image acquisition 396 Free-air ionisation chamber than 30 (sometimes 60) FPS. The frame rate must be taken into consideration when imaging moving anatomical regions such as the heart. Sometimes a term pulse rate is used instead of frame rate, but this is not correct for the imaging acquisition, as it could be related to the rate of x-ray pulses produced by the x-ray generator in sync with the image acquisition. Related Articles: Acquisition modes for digital image, Serial exposures, Pulsed mode Fraunhofer zone (Ultrasound) See Diffraction Free-air ionisation chamber Related Articles: Frame mode for digital image acquisition, (Radiotherapy) One of the most significant applications of ioni- Radionuclide imaging sation chambers is in the measurement of the dosimetric quan- Further Readings: Bushberg, Seibert, Leidholdt and Boone. tity exposure in a photon beam. An air-filled ionisation chamber 2012. The Essential Physics of Medical Imaging, 3rd edn., is particularly well-suited for this purpose because the exposure Lippincott Williams and Wilkins; Cherry, Sorenson and Phelps. is defined in terms of the amount of ionisation charge created in 2012. Physics in Nuclear Medicine, 4th edn., Elsevier. a mass of air. Due to the exposure definition, its measurement at a particular point requires the collection of all the ions of any Frame mode for digital image acquisition one sign generated in a known mass of free air contained in F (Diagnostic Radiology) See Acquisition modes for digital image a volume placed around that point. In practice, it is needed to follow over their entire range each secondary electron created Frame rate in the sensitive volume of the ionisation chamber by the pho- (Nuclear Medicine) The frame rate is the frequency by which an ton interaction and to measure all the ionisations created along imaging system produces new unique consecutive images. The their track. As the range in air of the produced secondary elec- frame rate is normally expressed in frames (or images) per second trons can be very large, it is impractical to design an instrument or in hertz (Hz). that permits a direct measurement and therefore a principle of compensation is used. If the sensitive volume of the ionisation Frame rate chamber is surrounded by a sufficient thickness of equivalent (Ultrasound) The frame rate is the nominal frequency at which air which is also subject to the same exposure, an exact com- an ultrasound image is updated. It applies to B-mode and colour pensation will occur as the ionisation charge created outside the flow imaging and is usually indicated on the screen image in units volume, from secondary electrons which were formed within of Hz or fps (frames per second). The time taken to image an area the volume, is

precisely balanced by the charge created within is dependent on the number of pulses used for each frame and the the volume from secondary electrons produced in the surround- pulse repetition frequency. This in turn depends on a number of ing air. This condition is called CPE. This situation is shown in factors, for example whether B-mode or colour flow imaging is Figure F.64. used, the depth and width of the image, line density and number of pulses used for each ‘line’ of ultrasound (e.g. if multiple trans- mit zones are used). It is possible that the colour flow and B-mode frame rate may differ in an image where both are displayed and updated. Frame rates typically range from 50 Hz for superficial B-mode images to 4 Hz for deep structures with a large area of colour flow imaging. Manufacturers go to considerable lengths to offset other parameters (e.g. line density) to ensure that frame rate remains acceptable for the application. Frames per second (Diagnostic Radiology) Frames per second is the number of images captured per second in procedures producing a continu- ous series of images. In analogy imaging, with video recorders or cineradiography the terms ‘video frame’ or ‘cine film cadre’ Volume were used. In digital fluoroscopy or radiography, the frames per second or (frame rate) are a measure of the temporal resolution of the imag- ing system (images acquired per second). Fast frame rates could leave a ghost image of the previous frame/image (afterglow) over the current frame/image and special methods are used to reduce this effect. Usually several sets of frames per second (FPS) exist in an equipment – e.g. 30, 15, 10, 4, 2. Although some digital FIGURE F.64 The ionisations produced by electrons outside the volume imaging equipment could have very high FPS (e.g. 1000 FPS in are balanced by ionisations within the volume which is produced by elec- industrial radiography), most anatomical uses do not require more trons originated elsewhere. Free induction decay (FID) 397 Free induction decay (FID) Signal Collimator Photon beam Sensitive volume FIGURE F.65 Schematic diagram of the free-air ionisation chamber. The design of the ionisation chamber based on this compen- sation is schematically shown in Figure F.65 and the ionisation TABLE F.2 F chamber is called free-air ionisation chamber. Air Thickness for Establishment of The incident radiation is collimated so that it passes between Electronic Equilibrium the electrode system of a parallel plate ionisation chamber. The system consists of a plate maintained at a high potential and a col- Photon Energy (MeV) Thickness (g cm−2) lecting plate surrounded by earthed guard rings. The guard rings — — are necessary to ensure that the electric field lines are perpendicu- 0.02 0.0008 lar to the electrodes in the sensitive volume space, avoiding lateral field distortion. 0.05 0.0042 To make all the secondary electrons generated in the sensi- 0.1 0.014 tive volume complete their tracks in air, the volume should be 0.2 0.044 surrounded by an air thickness greater than the range of the most 0.5 0.17 energetic secondary electrons. In practice, one prefers to mea- 1 0.43 sure the exposure (E) at the location where the entrance port is 2 0.96 given by 5 2.5 10 4.9 æ d¢ 2 ö q q E = ç = è d ÷ ø rA¢L rAL where in air would become very large. This would make the chamber d is the distance of the entrance port from the source very bulky and also the various required corrections and their d′ is the distance from any point in the source uncertainties would become problematic. q is the collected charge The free-air ionisation chamber requires correction for ambi- ρ is the density of air ent temperature and pressure. L is the length of the collecting plates Related Articles: Pressure and temperature correction factor A′ and A, respectively, are the cross-sectional area at a distance Further Reading: Knoll, G. 2000. Radiation Detection and d′ and d Measurement, John Wiley & Sons, Inc., New York. The distance from the entrance port to the sensitive volume Free induction decay (FID) of the ionisation chamber should be not so large to attenuate the (Magnetic Resonance) The application of a 90° excitation pulse photon beam significantly. rotates the net magnetism from the z to the y direction. The spins In Table F.2, the minimum air equivalent thickness required will then precess about the z direction in the transverse plan xy at to establish the electronic equilibrium condition is reported. The an angular frequency γB, where B is the field that the individual thickness is based on the range of secondary electrons in water. nuclear spins experience and γ is the gyromagnetic ratio. They The free-air ionisation chamber is the primary standard for will also simultaneously relax to alignment along the z direction. air kerma in air for superficial and orthovoltage x-rays produced If there is a current loop in a plane parallel to the x, z plane, the by applied voltage up to 300 kV. The free-air ionisation cham- changing magnetic flux in the loop will induce a current of fre- ber cannot function as a primary standard for photons beams of quency ω0 in this loop as shown in Figure F.66. higher energy because the air thickness surrounding the sensitive The resulting signal, an exponentially damped sine wave, is volume needed to establish the electronic equilibrium condition called FID because the spins begin to precess freely, the signal Free radicals 398 Frequency Z of the hydroxide ion and is highly reactive with a half-life in the body of around 10−9 s. Both types of free radical may participate in unwanted reactions resulting in cell damage, but it is estimated that around two-thirds of x-ray damage to DNA in mamma- Y lian cells is caused by the hydroxyl radical. Reactions between free radicals and DNA resulting in mutations have been linked to many forms of cancer. Free radicals have also been linked to many diseases, including Parkinson’s disease and Alzheimer’s, and to the aging process. Abbreviation: DNA = Deoxyribonucleic acid. X Further Readings: Hall, E. J. and A. J. Giaccia. 2006. Radiobiology for the Radiologist, 6th edn., Lippincott Williams & Wilkins, Philadelphia, PA; Koutsilieri, E. et al. 2002. Free radi- cals in Parkinson’s disease. J. Neurol. 249 (SupPlease 2):II/1–II/5; Pratico, D. et al. 1998. Increased F2-isoprostanes in Alzheimer’s disease: Evidence for enhanced lipid peroxidation in vivo. FASEB J. 12:1777–1783. Frequency FIGURE F.66 A magnetic moment rotating in the xy plane induces a (Ultrasound) The frequency ( f) of an ultrasound wave is the num- voltage in a coil in the yz plane. ber of compressions (or rarefactions) per second, Figure F.68. The unit for frequency (1/s) is Hertz (Hz). Ultrasound is a sound with a F frequency higher than 20 kHz. In diagnostic imaging, frequencies in the range of 2–20 MHz are most often used. High-frequency S ultrasound contributes to high-resolution images but reduces pen- etration depth as attenuation increases with higher frequencies. In modern scanners, the frequency of the transmitted pulses is +1 adjusted automatically depending on the transducer and the type of diagnostic option that have been chosen. A period (T) is the time it takes for a signal to repeat itself 0 and T = 1/f, Figure F.69. The acoustic wavelength (λ) is the dis- t tance between two corresponding points on the wave (e.g. two –1 Waves Compression FIGURE F.67 The free induction decay of a single signal. starts to decay with y time and the spins induce a current in the receiving coil. In Figure F.67, the FID signal is shown. The signal is of the form exp( - t /T * 2 )cos( - w0t), where T * 2 is Rarefaction the total transverse relaxation time. Two effects contribute to the decay rate 1/T * 2 . The first effect is the spin-lattice relaxation pro- cess representing the return of magnetisation to the z direction while the second effect is because the different nuclei in different environments and locations experience different fields differing by ΔB from the mean. T * 2 is always less than T2 and the relation between T * 2 and T2 is 1 / T * 2 = 1 / T2 + gDB. T * 2 also depends on dif- fusion, i.e. how rapidly spins spread out and leave the lattice. Pressure Free radicals (General) A free radical is an atom or molecule with at least one unpaired orbital electron in its outer shell. This results in a highly reactive state. In the body, free radicals usually occur on oxygen molecules and the two most important are the superoxide and hydroxyl radi- 1 cycle cals. Superoxide has the chemical formula O- 2 and is biologically toxic. Indeed, it is used by the immune system to kill invading FIGURE F.68 Definition of frequency, number of cycles per second. micro-organisms. The hydroxyl radical (OH) is the neutral form (Courtesy of EMIT project, www .emerald2 .eu) Frequency encoding 399 Frequency-tailored RF pulse Waves-basic terms by a negative gradient to dephase the spins so that they come back Period τ into phase at the centre of the read-out period, generating a gra- dient echo. (In a spin echo sequence, sequence timings are nor- mally such that the spin echo is also formed at this point in time.) Time This process of dephasing and rephasing is shown in Figure F.71, where the phase of transverse magnetisation at different locations along the gradient direction throughout the process is represented diagrammatically. Related Articles: B0 gradients, Phase encoding, Multi- Frequency 5 Hz, period 0.2 s slice, Slice selection, Readout period, Readout gradient, Image Period = 1 Frequency 5 MHz, period 0.2 μs reconstruction Frequency Further Reading: McRobbie, D. W. et al. 2007. MRI from Picture to Proton, Cambridge University Press, Cambridge, UK. FIGURE F.69 Definition of period and frequency. (Courtesy of EMIT project, www .emerald2 .eu) Frequency spectrum (Magnetic Resonance) A representation of a variable, for example compression peaks), and as the wave travels with speed of sound signal intensity as a function of frequency is denoted a frequency (c), the relationship between λ and f can be written as λ = c/f. spectrum. See also Fourier spectrum framework. Related Article: Wavelength Related Articles: Fourier spectrum framework, Fourier trans- form, Free induction decay Frequency encoding (Magnetic Resonance) Frequency encoding is one of the princi- Frequency-tailored RF pulse pal methods used for spatial localisation in MRI. It is normally (Magnetic Resonance) A frequency-tailored RF pulse contains F used in conjunction with phase encoding to localise signal within energy only within a specified frequency range. This is important a selected slice. for slice excitation or for frequency-selective suppression pulses. Frequency encoding is achieved by applying a static field The frequency range of the RF pulse can be tailored (designed) gradient while the NMR signal from the slice is being collected by manipulating the shape and the duration of the RF pulse. The (normally in the form of an echo). In the presence of this gradi- effect of an RF pulse on the spin population can be determined ent, elements of transverse magnetisation at different locations by the Bloch equations, and for the small flip angles, the Bloch along the direction of field variation precess at different frequen- equations can be approximated by the Fourier transformation of cies, according to the equation ω(x) = γB = γ(B0 + Gxx), where x the RF shape in the time domain. The possibility to design the is displacement along the gradient direction, Gx is the gradient frequency profile of RF pulses is frequently used in MRI to obtain amplitude (in mT m−1), γ is the gyromagnetic ratio of the nucleus a manifold of important effects such as the slice selection profile (normally 1H) and B0 is the strength of the static magnetic field. and selective excitation of fat tissue. The resulting signal can be decomposed using Fourier methods In the case of slice selection (see also Pulse sequence), the to determine the amount of signal at each frequency and hence RF pulse is combined with the slice-selective gradient pulse. The at each position, resulting in a projection through the sample slice thickness can be adjusted by changing the RF bandwidth (Figure F.70). (the selected frequency interval) and/or the amplitude of the In early imaging

techniques, acquisition was repeated with the slice-selective gradient. Slice-selective RF pulses have duration frequency encoding gradient at different angles, and the image of a few milliseconds (ms), and their shape is designed to create, was reconstructed using filtered backprojection. This approach ideally, a rectangular slice profile and homogeneous flip angles has recently come back into favour in certain applications, but over the whole width of the slice. Usually RF pulses with a sinc– two-dimensional Fourier imaging using frequency encoding and Gaussian shape are used for this purpose. phase encoding remains dominant. Related Articles: Flip angle, Pulse sequence, RF pulse In a conventional pulse sequence, the frequency encoding (or read-out) gradient is the last gradient to be applied and is preceded Gx ωhigh ωhigh NMR signal ωlow FT ωlow t FIGURE F.70 Frequency encoding generates a projection through the FIGURE F.71 Diagrammatic representation of frequency encoding gra- imaged object. dients and resulting phase evolution in a conventional pulse sequence. Fresnel zone 400 Fringe field Fresnel zone Fricke dosimeter (Ultrasound) See Diffraction (Radiotherapy) The standard Fricke dosimeter solution consists in 0.001M FeSO4 or Fe(NH4)2(SO4)2 and 0.8 N H2SO4 dissolved Fricke-based gel in air-saturated (∼0.250 mM O2) triple-distilled water. As the (Magnetic Resonance) The use of iron (II) sulphate solutions for concentrations are small, the dosimeter can be considered to be ionising radiation dosimetry has a lengthy history, having been effectively water. The conversion of ferrous to ferric ions induced originally proposed by Fricke in 1927. Irradiation of water gen- by the radiation permits the use of the Fricke dosimeter as a sec- erates free radicals, which convert Fe2+ ions into Fe3+ through a ondary standard. The yield of the ferric ions in the solution is variety of reactions. The quantity of Fe3+ ions produced depends directly proportional to the absorbed dose and is independent of linearly on the radiation dose received. In 1984, Gore proposed the dose rate up to 106 Gy s−1. The principal reactions that yield incorporating Fricke solution into a gel matrix, so that incident ferric ion are radiation produces a stable spatial distribution of Fe3+ ions. This distribution can be interrogated using MRI and the dose distribu- OH • +Fe2+ ® OH - + Fe3+ tion determined through the differential effect of Fe3+ and Fe2+ ions on water proton relaxation times. OH • +OH• ® H O + Fe2+ OH OH F 3 2 ® - 2 + • + e + The gel dosimetry system is usually based on gelatine or aga- rose (although other gels have also been used), which have the H • +O HO F 2 2 ® 2 • + e + ® Fe3+ + HO- distinct advantage of tissue equivalent response to ionising radia- 2 tion. The Fe2+ ions are normally supplied in the form of ferrous ammonium sulphate. Oxygenation is required to facilitate the free The ferric ion concentration can be measured by the change in radical reactions, and there is an extensive literature regarding the absorbance using a spectrometer at a wavelength of 304 nm at other possible additives to increase sensitivity. which the ferric ions absorb strongly and the ferrous ions do not F The change in relaxation times is due to the greater paramag- absorb at all. netic moment of the Fe3+ ions as compared to Fe2+. Assuming that The absorbed dose to the dosimeter solution is calculated from dipole interactions with the iron ions dominate water relaxation the equation and that there is fast exchange of water molecules between the vicinity of the iron ions and bulk water, the change in the spin-lat- DOD tice relaxation rate (ΔR1) is directly proportional to the increase in D = e × l × r ×G (Fe3+ Fe3+ ) concentration, and hence to dose. These changes in relaxation rate can be interrogated using any suitable T1-mapping technique. where ΔOD is the difference in optical density between the irra- The advantage of gel dosimetry over other dosimetric tech- diated sample and the blank, ɛ = 2187 L mol−1 cm at 304 nm niques is that it can visualise complex radiation fields with high and 25°C, l is the actual path length in centimetres through the spatial resolution over a large volume. Anatomically realistic spectrometer cell, ρ = 1.024 kg L−1 for standard Fricke solution at phantoms can be filled with gel, and inserts can readily be added 25°C and G(Fe3+) = 1.607 10−6 mol J−1 is the radiation chemical to mimic bone, lung etc. The Frick-based gel is itself the dosim- yield of ferric ions at the irradiation temperature (t) for Co60 γ eter, so the method yields a continuous dose distribution in three rays and varies slightly with radiation energy. A correction for dimensions without the need to introduce extraneous instruments. the temperature (t) differing from 25°C was made according to These advantages are particularly important for emerging con- the equation formal radiotherapy techniques where there may be abrupt varia- tions in dose, and also in brachytherapy. However, a number of complications have delayed the intro- eGt¢ = e25°CG25°C(1+ k1 (25 - t ))(1 + k2 (25 - t ’)) duction of Frick-based gels for routine dosimetric use. Careful attention to manufacture and calibration is needed to achieve where reproducible results. A dose of tens of Gy is generally needed to t′ is the irradiation temperature achieve significant changes in relaxation rate, with a fairly small t is the spectrometer measurement reading temperature region of linearity with dose before saturation at typically 50–75 Gy. Following irradiation, the imprinted dose distribution gradu- The temperature coefficient k1 is approximately 0.0007°C−1 and k2 is 0.0015°C−1 ally degrades due to diffusion of iron ions, so there is a window . of a few hours at most between irradiation and readout which The useful range of the Fricke dosimeter is from about 4 × 103 requires unusually ready access to MRI equipment. to 4 × 104 cGy. Some of these difficulties can be addressed through use of polymer-based gels, in which radiation-induced polymerisation Fringe field brings about changes in relaxation rates. These gels are free of (Magnetic Resonance) The magnet is one of the main compo- the diffusion problem associated with Fricke-based gels, but prob- nents of the MR scanner. The magnet is typically a large cylin- lems with reproducibility mean that they too are not yet in routine drical device that accommodates the patient during the MR. Its clinical use. role is to provide a highly uniform and stable polarising field in Related Article: Relaxation which to carry out the scan. There are several different types of Further Readings: Fricke, H. and S. Morse. 1927. The chemi- magnets: permanent, resistive and superconducting magnet. The cal action of Roentgen rays on dilute ferrosulphate solutions as a magnets used in MRI not only generate a field in the required measure of dose. Am. J. Roent. Radium Ther. Nucl. Med. 18:430– region but also produce a fringe field outside the magnet that 432; Gore, J. C., Y. S. Yang and R. I. Schulz. 1984. Measurement affect a considerable volume. The extent of the fringe field pro- of radiation dose distributions by nuclear magnetic resonance duced by a magnet depends on the strength of the magnet and on (NMR) imaging. Phys. Med. Biol. 29:1189–1197. its design. Permanent magnets are built of blocks of magnetic Fringe field 401 Fringe field material and can generate fields up to about 0.3 T. Their fringe devices. The effect of the stray field on ferromagnetic (‘magnetic’) field is often of limited extent and rather small. Resistive mag- materials is a torque and an attractive force and with electric cur- nets have a larger fringe field and internal fields up to about 0.15 rents is to produce a torque on the conductor carrying the current. T. Superconducting magnets represent the majority of manufac- A ferromagnetic object is attracted to the area where the magnetic tured magnets and can generate very high fields up to 10 T and field has the strongest gradient, i.e. to the region where the field more. They are a form of electromagnets which operate near is most inhomogeneous and not to the region where the absolute absolute zero temperature, i.e. 4.2°K. The fringe field of these field value is strongest. A long slender ferromagnetic object will types of magnet depends upon the current in the windings, the be strongly pulled into the top of the magnet and will fall freely dimensions and design of the coil system and the presence of any through the homogeneous region of the magnet and stick near the ferromagnetic material in the environment (Table F.3). At suf- bottom exit port of the magnet. The object will move freely back ficient distance, the fringe field (Bs) diminishes according to the through the homogeneous region of the magnet but will be very dipole approximation as the cube of the distance from the centre difficult to push out to the top. This is because, even though the of the magnet (r) and is proportional to the field strength of the field is stronger in the centre of the magnet, it has less magnetic magnet B0: gradient in the homogeneous centre of the magnet than near the ends of the magnet where the flux lines spread out on their indi- B vidual return path arcs. Obviously, the object will be a projectile Bs µ 0 r3 which falls into the magnet, and, therefore, the described experi- ence should not be tested to avoid severe damages to persons and Large distances are necessary, however, before the fringe field is equipment. reduced below the earth’s magnetic field (∼0.05 mT). The profiles of equal fringe field intensity when sectioned by a plane appear as a series of loops usually termed isogauss lines. For a typical –5 horizontal bore MRI magnet, these are illustrated in Figures F.72 through F.74, respectively, for 0.5, 1.0 and 1.5 T superconducting –4 F magnets on a horizontal plane across the isocentre (Table F.4). 0.1 mT Fringe fields can be substantially decreased through the use of 0.3 mT –3 0.5 mT a magnetic shielding. Shielding of magnets has the advantage of 1 mT reducing the controlled area required around the magnet room. 3 mT –2 Passive shielding is obtained applying highly permeable materials to walls or to the surface of the magnet maintaining the needed –1 homogeneity inside the bore. Active shielding is done by building a second magnet on the outside of the main magnet with a field 0 of the opposite sign, thereby reducing the fringe magnetic field (Table F.5). 1 Operation of numerous electronic devices in use in hospitals (e.g. x-ray tubes, CRTs, scintillation cameras, and image intensi- 2 fiers) may be affected by the fringe fields on the order of 0.1–5 mT. Effects on cardiac pacemakers have been reported for fringe 3 fields as low as 1.7 mT. The most common effect was triggering of the asynchronous mode; the effect is very model and orientation 4 dependent, and in the models tested, normal operation resumed when the pacemaker was removed from the field. Some pacemak- 5 ers also exhibited significant torque when exposed. For this rea- –4 –3 –2 –1 0 1 2 3 4 son, current static magnetic field guidelines restrict exposures for X, Y Metre wearers of cardiac pacemakers to 0.5 mT. This restriction is also applied to other implanted electronic devices and to prosthetic FIGURE F.72 Isohypses (contour lines) for a 0.5 T magnet. TABLE F.3 Characteristics of Different Types of Magnet Resistive Magnet Iron Characteristics Permanent Magnet Core Superconducting Magnet Field strength (T) 0.1–0.3 0.15–0.4 0.4–4 Distance to 0.5 mT fringe field (m) <1 0.5–2 3–10 Advantages Negligible fringe field Negligible fringe field High field strength Low operating cost Easy coil maintenance High field homogeneity Low capital cost Low capital cost Low power consumption Disadvantages Limited field strength Potential field instability Intense fringe field Fixed field strength High power consumption High cryogen cost Very heavy Water cooling necessary High capital cost Z Metre Front pointer 402 F ull field digital mammography –5 0.1 mT TABLE F.4 –4 0.3 mT Axial and Transverse Distances to Various Fringe 0.5 mT –3 1 mT Magnetic Fields for Several Magnet Strength, 3 mT Unshielded –2 0.5 T 1.0 T 1.5 T 4.0 T Fringe Axial/ Axial/ Axial/ Axial/ –1 Magnetic Transverse Transverse

Transverse Transverse 0 Field (mT) (m) (m) (m) (m) 1.0 7.0/5.0 8.5/6.7 10/7.0 13/11 1 0.5 8.5/6.7 11/8.5 13/9.7 16/13 0.3 11/8.3 13/10 15/12 20/16 2 0.1 15/12 19/15 21/17 25/20 3 4 5 TABLE F.5 –4 –3 –2 –1 0 1 2 3 4 X, Y Metre Methods to Reduce the Fringe Magnetic F Field FIGURE F.73 Isohypses (contour lines) for a 1.0 T magnet. Passive Shielding Active Shielding Iron in the wall Resistive shim coils Iron around the magnet Superconducting shim coils –6 0.1 mT Coil design –5 0.3 mT –4 0.5 mT 1 mT radiotherapy treatment beam onto the correct treatment location on the patient during treatment set-up. The pointer can be in the –3 3 mT form of a mechanical pointer or an optical pointer. –2 FSE –1 (Magnetic Resonance) See Fast spin echo (FSE) 0 Full field digital mammography 1 (Diagnostic Radiology) Digital mammography was first intro- duced commercially in 2000, taking advantage of digital tech- 2 nologies in order to overcome the drawbacks of traditional screen film mammography. 3 The term ‘full field digital mammography’ is also sometimes used nowadays for historical reasons to differentiate the ability to 4 image the whole breast instead of only small FOVs used initially for digital breast imaging. 5 In terms of x-ray tube and geometry, there are only minor dif- ferences compared to traditional screen film systems. However, 6 new anode filter combinations can now be used, such as W anode –4 –3 –2 –1 0 1 2 3 4 systems. X, Y Metre In screen film mammography, the film was the means for acquisition, processing, display and storage, while in FFDM, each FIGURE F.74 Isohypses (contour lines) for a 1.5 T magnet. one of these processes take place separately allowing more flex- ibility and room for optimisation. The image acquisition in FFDM is done through digital detec- Front pointer tors of low noise, wide dynamic range, linear or logarithmic (Radiotherapy) A front pointer is a device installed in the treat- response and relatively high spatial resolution (although lower ment head of a radiotherapy treatment machine, such as linear than the traditional film). accelerator for accurate positioning of the radiotherapy treatment The digital image can then be manipulated through post pro- beam. The pointer indicates the centre of the radiation beam cessing in order to enhance the visibility of lesions improving the emitted from a radiotherapy machine. It is used to direct the diagnostic value of the image. Z Metre Z Metre Full width at half maximum (FWHM) 403 Functional magnetic resonance imaging (fMRI) The digital nature of the image can furthermore offer capa- Pixel value bilities related to the use of computer aided detection systems, through simple algorithms or even sophisticated AI systems. Amax Full width at half maximum (FWHM) (Nuclear Medicine) This is a measure used to define the perfor- mance of a scintillation camera, in particular the spatial resolu- tion and energy resolution of an imaging system. Consider a point source placed in a detector FOV; the resulting image will have a wider spatial spread due to the properties of the camera sys- A FWTM tem. A count profile over the imaged point source is displayed max/10 in Figure F.75. The FWHM is the full width at half maximum, or Amax/2, expressed in the number of pixels or the distance in Pixel position (x, y) millimetres. According to NEMA, ‘The full width at half maximum is the FIGURE F.76 The FWTM is the curve width at one tenth of the maxi- measure of the spread of a point or line spread function mea- mum signal. sured between locations 50% down on each side from the peak amplitude’. Further Reading: National Electrical Manufacturers Functional magnetic resonance imaging (fMRI) Association. 2001. NEMA Standards Publication NU 1–2001, (Magnetic Resonance) fMRI is a technique which is used to study National Electrical Manufacturers Association, Rosslyn, VA, pp. brain areas which become active when the subject undertakes 8–10. a particular task. Whilst structural MRI is used to study brain anatomy, fMRI is used to study brain function. Historical Background: The development of fMRI during F Full width at one-tenth maximum (FWTM) (Nuclear Medicine) This is a measure used to define the scintilla- the 1990s was pioneered by Seiji Ogawa and Ken Kwong. Since tion camera performance (e.g. the spatial resolution, the amount then, fMRI has found clinical application in presurgical planning of scattered events and energy resolution of an imaging system). (to localise brain function). It has also been widely adopted as a Consider a point source placed in the FOV of a detector. The popular and insightful tool in the fields of psychology, psychiatry, resulting image will have a wider spatial spread due to degen- neurology and neuroscience. fMRI’s rapid growth over the past erative properties of the camera system. A count profile over the decade can be attributed to two main factors: it is a non-invasive imaged point source is displayed in Figure F.76. The FWTM is technique which is free from the risks associated with ionising the full width at one-tenth maximum, or Amax/10 showing the radiation and it has good spatial resolution (up to the order of a amount of scattered events in the curve. The broader the FWTM, millimetre). the larger the scattered events. Mechanism: fMRI is usually based on blood oxygen level– According to NEMA, ‘FWTM is the measure of the spread dependent (BOLD) contrast, which indirectly detects neuronal of a point or line response function measured between locations activity by imaging the accompanying increase in blood flow to 90% down on each side from the peak amplitude’. the local vasculature. The increased blood flow overcompensates Further Reading: National Electrical Manufacturers for increased oxygen consumption, such that the ratio of oxyhae- Association. 2001. NEMA Standards Publication NU 1–2001, moglobin to deoxyhaemoglobin rises in the active brain region. National Electrical Manufacturers Association, Rosslyn, VA, p. 2. Deoxyhaemoglobin is often referred to as a natural contrast agent: as a paramagnetic substance, it has a shorter T2* than oxyhaemo- globin (which is not paramagnetic, but diamagnetic). Thus, it is Full-wave rectification the changes in the deoxyhaemoglobin level which are imaged in (Diagnostic Radiology) See Rectifier fMRI. Brain Activation: During an fMRI study, subjects are imaged whilst simultaneously being presented with specific tasks or sen- sory stimulations which produce activity in certain regions of Pixel value their brain. Common examples of stimuli/tasks include Amax 1. Listening to sound tones for activation of the auditory cortex 2. Observing flashing lights (photic stimulation) for acti- vation of the visual cortex A 3. Finger/thumb motion for activation of the sensory max/2 FWHM motor cortex Scanning Parameters: Since deoxyhaemoglobin has a shorter T2* than oxyhaemoglobin, all fMRI images are T2* weighted. As gra- dient echo EPI (GE EPI) has a strong T2* weighting and rapid Pixel position (x, y) acquisition times, it is an ideal pulse sequence for fMRI although other techniques such as spiral imaging, conventional gradi- FIGURE F.75 The FWHM is the curve width at half of the maximum ent echo imaging and FSE imaging have also been used. Static signal. field strength is another important parameter: the BOLD contrast Functional MRI (fMRI) 404 FWTM (full width at one-tenth maximum) FIGURE F.77 Axial slices showing brain areas of activation during motor imagery (where the subject imagines performing a movement) in healthy controls. (Image courtesy of Stanton et al.) increases with increasing field strength. For this reason, higher basically of a thin wire of low-melting alloy which is heated by field strengths are often used for fMRI (Figure F.77). the electric current passing through it due to its electrical resis- Related Articles: Blood oxygenation level–dependent contrast tance. When the current exceeds the safe value for which the fuse (BOLD), Oxyhaemoglobin, Brain activation is designed, the wire either melts or vaporises, thus opening the Further Readings: Kwong, K. et al. 1992. Dynamic mag- circuit and stopping the current. The thin wire may be made of netic resonance imaging of human brain activity during primary aluminium, tin-coated copper or nickel. The fuse housing in elec- F sensory stimulation. Proc. Natl. Acad. Sci. USA 89:5675–5679; tronic equipment is most often a cylindrical glass or ceramic type Ogawa, S. et al. 1990. Brain magnetic resonance imaging with with a metal cap at each end. It is designed to resist the pressure contrast dependent on blood oxygenation. Proc. Natl. Acad. Sci. generated by the wire vaporisation, provided the voltage across the USA 87:9868–9872; Stanton, B. et al. 2007. Cortical activation fuse does not exceed its rating. There are two basic types of fuses: during motor imagery is reduced in amyotrophic lateral sclerosis. fast acting and slow blow. The fast-acting fuses blow very quickly Brain Res. 1172:145–151. when their particular current rating is exceeded. Slow-blow fuses have a coiled construction inside. They are designed to open only Functional MRI (fMRI) on a continued overload, such as a short circuit. The purpose of (Magnetic Resonance) See Functional magnetic resonance imag- coiled construction is to prevent the fuse from blowing on just a ing (fMRI) temporary current surge. A slow-blow fuse is usually used to pro- tect motors, and a fast-blow fuse to protect electronic equipment. Fuse (General) Fuse is a safety device used to protect an electric circuit FWHM (full width at half maximum) against excessive current. This is achieved by opening the elec- (Nuclear Medicine) See Full width at half maximum (FWHM) tronic circuit when current exceeds the amount, determined by the rating of the fuse. Opening a circuit under high current conditions FWTM (full width at one-tenth maximum) prevents equipment damages and overheating, which could cause (Nuclear Medicine) See Full width at one-tenth maximum a fire. A fuse is connected in series with the circuit and consists (FWTM) G Gadolinium-153 meglumine (an organic molecule containing ammonium, NH3). (Nuclear Medicine) A radionuclide used for quality assurance A class of nonionic agents was later developed (e.g. gadodiamide) purposes and in transmission sources for attenuation correction in which the metal-chelate complex is uncharged. These agents in gamma camera and SPECT imaging. have much lower osmolarity and from this perspective are argu- ably more suitable for in vivo use. Gadolinium chelate agents are usually administered by intra- Main venous injection, travelling in the circulation and freely crossing photon into the intravascular and intracellular spaces. Enhancement is emissions therefore noted in areas of enhanced vascularity, such as angio- Decay Daughter (Relative Common genesis associated with a tumour. They do not cross the intact Half-life modes nucleus intensity) application blood–brain barrier, leading to prominent enhancement in the 240.4 hr Electron Europium-153 97 keV Quality presence of a lesion causing disruption of this barrier (Figure G.2). (10) capture (29%) assurance / These agents have a range of clinical applications, including days (100%) 103 keV Transmission characterisation of tumours, visualisation and dating of cerebral (21%) sources ischaemia, and assessment of areas of inflammation and infec- tion. There is a growing area of clinical applications in cardiac imaging, including assessment of myocardial perfusion and of Gadolinium-153 has a relatively long half-life of 240 days and viability following infarction. main photon emissions around 100 keV. Owing to those proper- The superior safety profile of gadolinium chelates relative ties it has been or is currently being used for several applications to x-ray contrast agents, which can be nephrotoxic, combined G in nuclear medicine: with the increasing range of clinical indications for MRI, led to widespread use of these agents, including in patients with renal • Attenuation correction using moving line sources for problems. In some cases, such as for MR angiographic studies, gamma camera/SPECT imaging (sequential emission/ multiple doses of contrast agent were used. In 2006, an asso- transmission imaging). ciation was found between administration of gadolinium-based • Quality assurance measurements. contrast agents and a condition known as nephrogenic systemic • Dual photon absorptiometry for measuring bone min- fibrosis (NSF) in patients with severe renal impairment. The eral content. detailed mechanism remains under investigation at the time of • Gadolinium-153 microspheres for measuring coronary writing (2009), but putatively the increased biological half-life blood flow and perfusion. of gadolinium agents in the case of renal dysfunction increases time for release of free gadolinium, which may be the trigger Related Articles: Attenuation correction in SPECT using for NSF. Theoretically, ionic cyclic agents would be expected transmission scans,

SPECT (Single photon emission computed to be the most stable and nonionic linear agents the least stable, tomography), Quality assurance and this appears to match the pattern of NSF incidence. Careful Further Readings: Brookhaven National Laboratory. NNDC. screening for renal dysfunction, use of more stable agents and www .nndc .bnl .gov /nudat2/, last accessed 31 October 2019; less use of multiple doses have virtually eliminated new cases Cherry, Sorenson and Phelps. 2012. Physics in Nuclear Medicine, of NSF. 4th edn., Elsevier; Sharp, Gemmell and Murray. 2005. Practical Related Articles: Contrast agent, Paramagnetic contrast Nuclear Medicine, 3rd edn., Springer. agents, Positive contrast media Gadolinium chelate Gadolinium orthosilicate (GSO) (Magnetic Resonance) Gadolinium chelates are by far the most (Nuclear Medicine) GSO is an inorganic scintillation crystal commonly used form of contrast agent in MRI. They are a form which is used due to its relatively high atomic number (Z = 64). of paramagnetic contrast agent, usually used as a positive, T1- The crystal can be grown into reasonably large sizes, which is shortening agent. The dominant position of gadolinium is attrib- another attractive feature. The light yield and the decay time of utable to its large relaxivity (having seven unpaired electrons) and GSO depend on the level of doping, i.e. the amount of impurities large number of coordination sites (nine). added to the crystals. The maximum light yield is observed when A gadolinium chelate consists of a Gd3+ ion combined with a the doping concentration of cerium is about 0.5 mol%. The decay chelate molecule, the purpose of which is to bind the ion which time is dominated by one component (45 ns) but there is also a is highly toxic in its free form, causing rapid liver necrosis. A longer decay time of 400 ns with a 10% yield. The total light yield number of agents are commercially available, each based on a is almost 20% of the NaI(Tl) light yield. The energy resolution different chelate molecule (Figure G.1). is approximately 9% for 662 keV gamma rays which make GSO Gadopentetate dimeglumine was the first agent to become suitable for PET imaging systems. GSO has excellent radiation commercially available. The metal-chelate complex carries a stability; even after exposure of up to 107 Gy no appreciable radia- net negative charge, requiring neutralisation by the addition of tion damage effects have been reported. 405 Gadolinium oxysulphide 406 Gafchromic film CO2– –O2C CO2– N N –O2C CO2– N N N N N OH OC CO –O2C N N CH3 H CH3 H CH3 ProHance (gadoteridol) Omniscan (gadodiamide) –O2C CO2– CO N N 2– –O2C CO2– N N N N N –O2C CO2– –O2C CO2– Dotarem (gadoterate meglumine) Magnevist (gadopentetate dimeglumine) FIGURE G.1 Examples of commercially available gadolinium chelate structures. peak at 545 nm. GOS screens are used as intensifier screens (enhancing the light output) in x-ray cassettes used in screen- G film mammography. Related Article: Inorganic scintillators Gadopentetate dimeglumine (Gd-DTPA) (Magnetic Resonance) See Gadolinium chelate Gafchromic film (Radiotherapy) Gafchromic film is the popular vendor name of the self-developing radiochromic film (i.e. a film undergoing a change in colour corresponding to exposure to radiation) used for film dosimetry in various areas of medical physics (radiotherapy, diagnostic radiology, etc.). The sensitivity and the thickness of the emulsion vary depending on the use. The most common Gafchromic film is a light blue coloured film, has an almost tissue-equivalent composition and develops a darker blue colour when exposed to radiation. Other films are also available such as a yellow coloured film which develops a black colour when exposed to radiation. This colouration results from the polymerisation of a dye. The polymer will absorb light and so by using a densitometer is it possible to measure the transmission FIGURE G.2 Rim enhancement of a tumour disrupting the blood–brain of light through the film and relate this to the level of radiation barrier. exposure. With careful handling and calibration of the film it should be possible to achieve a precision of better than 3% for dosimet- Related Articles: Inorganic scintillators, Scintillators, NaI(Tl) ric work. The film is becoming increasingly popular in general detector crystal, Bismuth germanate (BGO) use due to the loss of film processing facilities in hospitals and Further Reading: Knoll, G. F. 2000. Radiation Detection and there has also been a reduction in the cost of radiochromic films. Measurement, 3rd edn., John Wiley & Sons, New York, p. 243. Gafchromic film is commonly used in brachytherapy and stereo- tactic radiotherapy due to its high spatial resolution and response Gadolinium oxysulphide for high doses, as well as for IMRT verification and general linac (Nuclear Medicine) Gadolinium oxysulphide (Gd2O2S) is QA measurements. an inorganic scintillator. When used in medical imaging it is Abbreviations: IMRT = Intensity modulated radiation therapy doped with terbium. Gadolinium oxysulphate is commonly and QA = Quality assurance. abbreviated as GOS. The wavelengths emitted in the scintilla- Related Article: Radiochromic film tion process range from 382 to 622 nm with a primary emission Hyperlink: www .gafchromic .com Gain 407 Gallium-67 [67Ga] Gain (Nuclear Medicine) The gain of electrical/electronic circuits is defined as the ratio between the output and the input signal. When talking about gain in nuclear medicine one often refers to the photomultiplier (PM) tube signal gain. A photomultiplier tube has a gain up to 108 depending on the design of the PM tube. Related Article: Photomultiplier (PM) tubes Gain (Ultrasound) In ultrasound imaging, gain is a control of the amplification of ultrasound signals. Gain can be applied to B-mode, M-mode, colour flow and spectral Doppler signals. It is an important control in everyday operation of scanners. The effects in ultrasound scanners are: B-mode: Increased gain makes the overall image appear brighter, conversely lowering the gain reduces all the levels used in the image (Figure G.3). The gain control modifies grey levels throughout the image, this can be modified by the depth/time gain control DGC/TGC where gain at different depths can be altered. A common fault for new users is to use too high a gain, thereby restricting the number of greyscales displayed. Colour flow imaging. Gain increases the amount of colour flow signal displayed. Too high a gain produces colour signal from tissue, too low gain risks missing colour signals in blood. PW spectral Doppler: Increasing gain increases signals from weak echoes but background noise is increased (Figure G.4). Increased gains will accentuate spectral broadening artefacts FIGURE G.4 High gain in pulsed wave spectral Doppler mode (lower G and cause elevated velocities to be measured. Low gains will image) causes noise to appear in the sonogram and increased measured peak velocity from 75 to 84 cm/s. reduce the intensity of the sonogram and lead to lower velocity measurements. One difficulty in selecting gain is that there is no common standard for measuring and displaying gain, some scanners show gain as dB, others as a number. This can lead to difficulties when comparing images or when users encounter a new scanner. Related Articles: Time gain control, Depth gain control, Sonogram, Colour flow imaging, B-mode, M-mode Gain, amplification (General) See Amplification factor Gain correction (Nuclear Medicine) Gain correction refers to compensations for non-linearities in electric charge of the multiplier response. The goal of gain correction is that every pixel should yield identical output signals for a given input signal and that this relationship is true for all input intensities. The gain can also differ between two individual photomultiplier (PM) tubes. Such unwanted difference in gain is corrected by a uniformity correction. Abbreviation: PM tubes = Photomultiplier tubes. Related Articles: Photomultiplier (PM) tubes, Uniformity FIGURE G.3 Image of a kidney (top) and (lower) the same view with Gain factor the gain increased by 17 dB. With high gain the whole image appears (Nuclear Medicine) The factor by which an electrical signal is brighter but dark grey levels are not used in the image and contrast is magnified in an electronic charge multiplier, e.g. a photomulti- reduced and detail lost. plier tube. Gallium-68 [68Ga] 408 Gamma camera The gain factor in a photomultiplier (PM) tube used for nuclear W. Falen. 2004. Radiopharmaceuticals in Nuclear Pharmacy and medicine imaging depends on PM tube design, but can be as high Nuclear Medicine, 2nd edn., American Pharmacists Association, as 108. Washington, DC. Related Article: Photomultiplier (PM) tubes Gallium-68 [68Ga] Gallium-67 [67Ga] (Nuclear Medicine) (Nuclear Medicine) Element: gallium (group III-A) Element: gallium (group III-A) Isotopes: 56 < N < 86 Isotopes: 56 < N < 86 Atomic number (Z): 31 Atomic number (Z): 31 Neutron number (N): 37 Neutron number (N): 36 Symbol: 68Ga Symbol: 67Ga b+, EC Production: Generator 68 Zn(p,n)68Ga ® 68 Zn Production: Cyclotron, e.g. 65 EC Cu(a,2n)67Ga ® 67 Zn 67.63min Daughter: 68Zn Daughter: 67Zn 3,26d Half-life: 67.63 min Half-life: 3.26 days Decay mode: β+ (89%) and EC (11%) Decay mode: EC-decay Radiation: β+ (max 1899 keV), annihilation photons Radiation: gamma, internal conversion electrons, Auger Photon energy: 511 keV (178%) electrons, characteristic x-ray photons Dose rate from 1 MBq: 0.140 μSv/h at 1 m; 1400 μSv/h at 1 cm Gamma energy: 93.31 keV (39.2%), 184.58 (21.2%), 300.21 keV Absorption (average range of β+): 2.15 mm in tissue (16.8%), 393.53 keV (4.7%) Critical organ: small intestine, upper large intestine, red bone Dose rate from 1 MBq: 0.0208 μSv/h at 1 m; 208 μSv/h at 1 cm marrow, bone surfaces, spleen and adrenals Absorption (HVL): 1 mm lead ALImin (50 mSv): 600 MBq Biological half-life: 12 years (citrate) Absorbed dose (gallium citrate): 0.064 mGy/MBq small intestine, Critical organ: bone surfaces, lower large intestine, red bone mar- 0.053 mGy/MBq upper large intestine, 0.046 mGy/MBq red row, spleen, adrenals. bone marrow, 0.037 mGy/MBq bone surfaces, 0.036 mGy/ ALImin (50 mSv): 300 MBq MBq spleen, 0.034 mGy/MBq adrenals. G Absorbed dose (gallium citrate): 0.59 mGy/MBq bone surfaces, Effective dose: 0.027 mSv/MBq 0.20 mGy/MBq lower large intestine, 0.19 mGy/MBq red bone marrow, 0.15 mGy/MBq spleen, 0.14 mGy/MBq adrenals. Effective dose: 0.12 mSv/MBq 31 2P1°/2 1+ 67.629 m 67 31 2P Ga Ga 1°/2 31 EC Gallium Ga 3/2– 3.2612 d 69.723 Gallium 67 [Ar]3d10 4s2 4p EC 31Ga 69.723 5.9993 QEC 2921.1 [Ar]3d10 4s2 4p 5.9993 Clinical Applications: Its clinical use is so far limited, but theoretically the same as for the gallium isotope 67Ga, i.e. for non- Clinical Applications: In contrary to indium, which is specific imaging and/or localisation of non-Hodgkin’s disease, labelled to a number of pharmaceuticals, the only existing radio- lymphoma, bronchogenic carcinoma, and inflammatory lesions pharmaceutical made by gallium is 67Ga-citrate. It is available as a (e.g. sarcidosis) or fever of unknown origin using PET. sterile aqueous solution, in most cases with an activity of 37 or 74 Because of a radioactive equilibrium between 68Ge and 68Ga, MBq in 1 mL. 67Ga-citrate is prepared by neutralising acidic NCA 67 the first-mentioned nuclide is used as standard calibration source Ga-chloride with sodium hydroxide with 4% sodium citrate. The for PET cameras. injection solution has pH 5.5–8. The recommended administered Related Article: Gallium-67 activity is 74–185 MBq. 67 Further Readings: Annals of the ICRP. 1987. Radiation Dose Ga-citrate can be used clinically for non-specific imaging to Patients from Radiopharmaceuticals, Biokinetic Models and and/or localisation of non-Hodgkin’s disease, lymphoma, and Data, Vol. 18, ICRP Publication 53, Pergamon Press, Oxford, bronchogenic carcinoma. 67Ga-citrate is also used for the localisa- UK; Annals of the ICRP. 1998. Radiation Dose to Patients tion of inflammatory lesions (e.g. sarcidosis) or fever of unknown from Radiopharmaceuticals, Vol. 28, No.3, Addendum to ICRP origin. Publication 53, ICRP Publication 80, Pergamon Press, Oxford, Related Articles: Indium-111, Gallium-68 U.K.; Firestone, R. B. 1999. Table of Isotopes, 8th edn., Update Further Readings: Annals of the ICRP. 1987. Radiation Dose with CD-ROM [http://ie .lbl .gov /toi .html]; Kowalsky, R. J. and S. to Patients from Radiopharmaceuticals, Biokinetic Models and W. Falen. 2004. Radiopharmaceuticals in Nuclear Pharmacy and Data, Vol. 18, ICRP Publication 53, Pergamon Press, Oxford, Nuclear Medicine, 2nd edn., American Pharmacists Association, UK; Annals of the ICRP. 1998. Radiation Dose to Patients Washington, DC. from Radiopharmaceuticals, Vol. 28, No.3, Addendum to ICRP Publication 53, ICRP Publication 80, Pergamon Press, Oxford, UK; Firestone, R. B. 1999. Table of Isotopes, 8th edn., Update Gamma camera with CD-ROM [http://ie .lbl .gov /toi .html]; Kowalsky, R. J. and S. (Nuclear Medicine) See Scintillation camera Gamma camera SPECT systems 409 Gas amplification Gamma camera SPECT systems (Nuclear

Medicine) See Single photon emission computed tomog- raphy (SPECT) Gamma correction tables (Diagnostic Radiology) Gamma correction tables are rarely used today. These are necessary in radiography when the type of x-ray film is changed (a new film with different gamma or latitude is used). The tables show the necessary change of some of the basic radiographic parameters (kV, mA s) which will present over the new film the usual anatomical structures with the same optical densities (gray levels). The tables are made for each anatomical structure and each x-ray equipment in the department. The tables are usually made by the respective senior radiographer and medi- cal physicist. In digital radiography similar adjustment (present- ing identical pixel values for identical structures, while x-raying with the same radiographic parameters) is usually made by the service engineers during the system maintenance. Related Articles: Exposure point, Film latitude Gamma knife (Radiotherapy) The gamma knife is a radiotherapy device for treating brain tumours using external-body radioactive sources originally developed by Leksell. A shielded applicator is placed around the patient’s head. Up to 201 cobalt 60 sources are placed inside the applicator with the shielding used to collimate the dose FIGURE G.5 Radiotherapy machine showing gantry support for the to deliver a high intensity to a target point in the patient’s brain. radiation source. G The gamma knife is used to deliver radiosurgery, i.e. a single frac- tion treatment. The applicator is affixed to the patient’s skull to ensure it remains fixed relative to the treatment target. This con- The gantry (Figure G.5) is the part of the radiotherapy or tains the collimation system for the sources. imaging equipment supporting the radiation source and the col- Related Articles: Stereotactic, Radiosurgery, Stereotactic frame limation system and allowing their movements. The gantry can be moved in a circular path permitting the sources of radiation to Gamma radiation rotate on about a horizontal axis (gantry axis). During the rota- (General) Gamma radiation is one type of electromagnetic radia- tion of the gantry the beam axis and the collimator axis move in a tion, characterised by its range of energy and its mode of pro- vertical plane around a point called the isocentre. duction. It has no upper limit to its energy and the lower limit Related Articles: Isocentre, Radiation isocentre, Mechanical is approximately 0.1 MeV. It is produced as a result of nuclear isocentre transformations. At the lower end of its energy range it overlaps, in terms of Gas amplification energy, with x-rays. (Radiation Protection) The gas amplification in gas-filled radia- Gamma radiation and gamma rays are synonymous terms. tion detectors depends not only on the kind of gas and its pres- Related Article: Nuclear transformations sure but also on the voltage between electrodes. In Figure G.6 a scheme of the measured current versus potential between the Gamut electrodes for x- or gamma radiation is shown. (Diagnostic Radiology) Gamut is a term describing a complete The gas amplification occurs when the primary electrons range of things (e.g. set of colours, musical notes). The colour (e.g. photoelectrons, Compton electrons produced by x- or gamut of a colour monitor is the complete subset of colours which gamma radiation) and ions accelerated by the potential between it reproduces/visualises, compared with the whole range of pos- the electrodes gain sufficient energy to ionise the gas and pro- sible such colours in the chromaticity diagram – i.e. the visual duce secondary electrons that can create further ionisation. The gamut of any monitor is limited. This is a specific characteristic electric charge Q created by n0 primary electrons and ions is of the colour monitor. equal: Hyperlink: https://en .wikipedia .org /wiki /Gamut Q = n0 e M Gantry (Radiotherapy) The gantry is the part of the radiotherapy or imag- where e = 1.6 × 10−19 ing equipment supporting the radiation source and the collimation C system and allowing their movements. The gantry can be moved M is the average gas amplification factor in a circular path permitting the sources of radiation to rotate on about a horizontal axis (gantry axis). During the rotation of the The value of gas amplification is proportional to the applied gantry the beam axis and the collimator axis move in a vertical voltage and depends on the detector dimensions. plane around a point called the isocentre. Related Article: Proportional counter Gas flow counters 410 G ate Gas End- 4 window 3 + 2 – R 1 Current FIGURE G.8 Scheme of a cylindrical gas-filled radiation detector. 0 Voltage FIGURE G.6 Scheme of the measured current as a function of the Gas-filled radiation detectors potential between the electrodes in a gas-filled detector for x- or gamma (Radiation Protection) In gas-filled radiation detectors the ion- radiation: (1) ionisation chamber region, (2) region of proportional coun- ised gas molecules and electrons are collected by electrodes ter, (3) region of Geiger counter, (4) continuous discharge. (Figure G.8) as current or charge due to the applied voltage. The quantity of current or charge depends not only on the kind of gas and its pressure but also on the voltage between elec- Further Reading: Knoll, G. F. 2000. Radiation Detection and trodes. In Figure G.6 the variation of the measured current ver- Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. sus potential between the electrodes for x- or gamma radiation 169–173. is shown. Photons (x or gamma) passing through gas can interact with Gas flow counters its molecules by the photoelectric or Compton effects. If their G (Radiation Protection) Gas flow counters are used for small sam- energy is equal or higher than 1.022 MeV they can produce ples emitting alpha and beta particles or soft photon radiation. electron-positron pairs. The electrons resulting from these pro- The radioactive sample is introduced in the gas-filled detector cesses are accelerated by the applied voltage between the elec- (Figure G.7.) operating in the proportional region (see the follow- trodes and have enough energy to ionise the gas. In ionisation ing Related Articles). The gas flows through detector during the chamber region (1) all electrons and ions are collected by the counting. The counting efficiency is almost 100%. electrodes. The measured current (charge) is proportional to Usually noble gases in binary mixtures are used (e.g. a mix- the initial energy of radiation. If the applied voltage increases ture of 90% argon and 10% methane) to ensure good value of the the electrons and ions acquire increased kinetic energy and gas multiplication factor. can produce additional ionisation which gives a multiplication If the aim of the studies is an evaluation of the deposition of of collected charge (multiplication in range 102–104). This is ionising radiation energy in biological tissue, it is necessary to use the proportional counter region (2), the current is proportional a gas mixture consisting of 64.4% methane, 32.4% carbon dioxide to the initial quantity of radiation (initial number of ions). At and 3.2% nitrogen. higher voltage is the region of limited proportionality. Then we The same technique can be used in flow Geiger–Müller coun- pass to the Geiger region (3) where an avalanche of interactions ters. In a flow Geiger–Müller detector the sample, if it is in the is generated and relatively large gas amplification occurs. In form of gas, can be mixed with the counter gas (argon-alcohol). a Geiger region of operation the measured signal (current or This method is applied to detect low concentration of C-14 intro- charge) is independent of the initial energy of radiation. The duced in the form of CO2 and H-3 added as a vapour of the triti- number of ion pairs produced in one discharge is about 109– ated water. 1010 and the amplitude of signals is so large that (its) ampli- Related Articles: Gas-filled radiation detectors, Geiger– fication is not used. Further increase of the voltage produces Müller (GM) counters, Proportional counter spontaneous discharge which may damage the detector. This is Further Reading: Knoll, G. F. 2000. Radiation Detection and the (a) continuous discharge region (4). Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. Related Articles: Geiger–Müller (GM) counters, Ionisation 165, 211. chamber, Proportional counter Further Readings: Delaney, C. F. G. and E. C. Finch. 1992. Radiation Detectors, Oxford University Press, Oxford, New York, pp. 66–107; Dendy, P. P. and B. Heaton. 1999. Physics for Diagnostic Radiology, 2nd edn., Institute of Physics Publishing, Gas R inlet Philadelphia, PA, pp. 139–141; Hobbie, R. K. 1997. Intermediate Sample Physics for Medicine and Biology, 3rd edn., Springer-Verlag, Gas + New York, p. 424; Saha, G. P. 2001. Physics and Radiobiology outlet – of Nuclear Medicine, 2nd edn., Springer-Verlag, New York, pp. 66–74. Gate FIGURE G.7 Scheme of a gas flow counter with a radioactive sample (Ultrasound) Gate is a term used in pulsed wave Doppler denot- inside. ing the range over which Doppler investigation is conducted along Measured current (signal) Gated SPECT 411 Geiger–Müller (GM) counters the Doppler beam. Gate is also described as range gate or sample reflective markers on the patient’s surface, a pressure sensitive belt, volume. spirometry and x-ray imaging, (another approach is developed by See Sample volume a company using fiducial makers implanted near the diaphragm. Related Articles: Sample volume, Pulsed wave Doppler These markers are tracked by x-ray imaging – CF Brainlab). Abbreviations: ABC = Active breathing control, CT = Gated SPECT Computed tomography and DIBH = Deep inspiration breath hold. (Nuclear Medicine) Gated SPECT imaging makes it possible for Related Articles: Respiratory gating, Active breathing con- a temporal display of the perfusion distribution of the myocar- trol, Deep inspiration breath hold technique dium during a heart cycle. Gated myocardial perfusion SPECT imaging is an acquisition synchronised to the patient’s electro- Gauss cardiogram (ECG). A number of projection images (8 or 16, also (General) Gauss (G) is a c.g.s unit for magnetic flux density B called intervals or frames) are acquired at each projection angle, (also magnetic induction): with each image or frame corresponding to a specific part of the cardiac cycle. For maintaining adequate count density of the indi- 1G = 10-4 T vidual cardiac frames a high count density, often 99mTc-labelled tracer is used. All projection images for a given cardiac inter- val are reconstructed separately to a SPECT image volume. The (c.g.s. – the centimetre, gram second system of units, the c.g.s is SPECT volume corresponding to the various cardiac intervals can now replaced by SI). be viewed in 4D format allowing for evaluation of dynamic car- One gauss is defined as one maxwell per square centimetre. diac function. In the SI system magnetic flux is measured in weber (Wb), The determination of the myocardial mass, the endocardial, thus magnetic flux density of 1 T is the epicardial surface and the valve plane for each gating frame allows for calculation of the end-diastole and the end-systole 1T = 1Wb/m2 (weber persquaremetre) volume from which the ejection fraction can be derived. Gated SPECT images thus allow for evaluation of both perfusion and The gauss is named after the German physicist (and mathemati- function of the myocardium. The extent and severity of the perfu- cian) Karl Friedrich Gauss. sion defect, left ventricular ejection fraction as well as regional Related Article: Tesla and global ventricular functions can be assessed. G Related Articles: Severity, Extent Gaussian distribution (General) The Gaussian distribution, or the normal distribution, Gating, cardiac is a continuous probability distribution that has a mean value and (Magnetic Resonance) See Cardiac gating a variance. The standard Gaussian has a mean of zero and a vari- ance of one. The shape of the distribution looks very much like a Gating, respiratory bell that falls rapidly towards plus/minus infinity. (Radiotherapy) The equation for the Gaussian distribution is given by Imaging Gating: Gating may be used in radiological imaging to reduce the artefacts due to the effects of organ motion during 2 the cardiac and respiratory cycles. In imaging for radiotherapy, 1 æ ( x - m) ö expç - ÷ s 2p ç 2s2 the respiratory cycle is particularly problematic as a treatment ÷ irradiation typically lasts one or several minutes, encompassing è ø several breathing cycles, during which internal anatomy, includ- where ing the tumour, may move significantly. Gating may be used to μ is the mean acquire a snapshot scan at a particular phase of the cardiac or σ is

the standard deviation (variance) breathing cycle. An emerging use of gating is so-called 4D acqui- sition, particularly in CT. A scan is taken of the patient over sev- Gaussian functions (or filters) are widely used in image pro- eral breathing cycles and each data point is labelled with phase of cessing to characterise the blurring effect of scintillation camera the cycle. The data are then reconstructed to yield several scans collimators. covering several phases throughout the cycle. Treatment Gating: Gating of treatment is used in external Gaussian noise beam radiotherapy to minimise the effects of motion, specifi- (General) Gaussian noise is noise whose probability distribution is cally breathing motion, on the treatment accuracy. This technique that of the Gaussian distribution (normal distribution). Gaussian makes it possible for the clinician to define a smaller target volume noise is equal to the standard deviation, σ, in the Gaussian distri- so as to protect the organ at risk (the lung) and increase dose to bution equation. the target volume). Two essentially different approaches exist: to Related Articles: Gaussian distribution, Poisson noise, gate the patient’s breathing using active breathing control (ABC) Poisson distribution or deep inspiration breath hold (DIBH); to measure the patient’s breathing and to gate the beam delivery by turning it on and off depending on the phase and/or position in the breathing cycle. GE Issues associated with gating the treatment include the duty cycle (Magnetic Resonance) See Gradient echo (GE) (i.e. the fraction of time the radiation beam is being delivered) and the characteristics of the beam as it is repeatedly turned on and off Geiger–Müller (GM) counters by the gating process. A variety of methods exist to measure the (Radiation Protection) The Geiger–Müller counter (G–M counter position in the breathing cycle, including the use of one (or several) or Geiger tube) was invented in 1928 by Hans Geiger and Walther Gel 412 Gel To high voltage supply Thin Anode G–M counter R end-window HV supply V(t) Anode Signal Insulator Cathode Cathode FIGURE G.9 Scheme of Geiger–Müller counter with thin end-window. FIGURE G.11 Scheme of the G–M counter electronic system. Müller. It is a gas-filled detector based on ionisation. The construc- tion is similar to ionisation chambers and proportional counters but Sons, Inc., New York; Saha, G. P. 2001. Physics and Radiobiology it works on the principle of gas multiplication. As a result all pulses of Nuclear Medicine, Springer, New York. from a G–M counter are of the same amplitude whatever the energy of the incident radiation. A diagram of a G–M counter is shown in Gel Figure G.9. The hollow metal tube (or glass tube with mesh cylin- (General) A gel is a crosslinked substance which exhibits thix- der) is negatively charged and filled with argon or neon gas at a low otropy, i.e. it is a solid at rest and becomes fluid when agitated. pressure. There is a positively charged electrode along the centre of A gel is mostly fluid, but behaves as a solid due to the networked the tube. X-ray, gamma photons or other particles (if the end-win- structure within the liquid. The density of a gel is similar to its dow is sufficiently thin) entering the counter cause ionisation of the constituent liquid. The crosslinking contributes to a gel’s hard- gas. The potential difference between electrodes is several hundred ness and toughness, which can vary considerably. Many fluids can volts. Electrons released as a result of the ionisation are attracted be used to form a gel including water (hydrogels), oil (organogels) to the positively-charged central anode. As they pass through the and air (aerogels). gas they are accelerated, gaining sufficient energy to eject electrons Gels are extremely versatile in their uses, ranging from foods G from other gas atoms. These new electrons are also accelerated, and to adhesives. They are used for fibre optic communication and to may cause further electrons to be released – a chain reaction. One prevent water intrusion and mechanical damage. A cationic poly- gamma photon can produce about 105 electrons. Quenching of the mer is often used in gels because its positive charges prevent coil- gas multiplication process is necessary to limit the amount of detec- ing, allowing the polymer to increase the gel’s viscosity. These tor dead time. The quenching is accomplished with use of a small polymers are used in hair gels as they bind to the anionic amino quantity of alcohol vapour in the gas. acids on the keratin molecules in hair. A graph of the operating characteristics of a gas counter is Hydrogels contain natural or synthetic water-insoluble poly- shown in Figure G.10. The region of the graph that a G–M coun- mers with hydrophilic groups, for example acrylates and agarose. ter operates in is region b, the plateau region corresponding to They are very absorbent and flexible, similar to biological tissue, the maximum gas amplification. The plateau is between 100 and due to their substantial water content. They can be made to be 1500 V. It depends on the size of the G–M counter. sensitive to changes of pH, temperature, or concentration of a In Figure G.11 the scheme of the G–M counter electronic sys- metabolite, causing them to swell as a result. Hydrogels are also tem is shown. used in disposable diapers (nappies) to absorb urine. Geiger–Müller detectors are used for detecting the presence of Organogels are non-crystalline thermoplastic solids in organic radiation, e.g. as a contamination monitor. liquids such as vegetable oils. The solubility and particle size Further Readings: Dendy, P. P. and B. Heaton. 1999. Physics determine the elastic properties and hardness of an organogel. for Diagnostic Radiology, IOP Publishing, Bristol, UK; Knoll, G. Organogels are used in pharmaceuticals, cosmetics, art conserva- F. 2000. Radiation Detection and Measurement, John Wiley & tion and food. Xerogels are solids formed from gels by drying them with unconstrained shrinkage. They maintain a very low density, a high porosity and a large surface area. Formation under hyper- critical conditions minimises shrinkage and decreases the gel’s density to produce an aerogel. Conversely, heat treatment causes (c) viscous sintering, thereby maximising shrinkage to produce a (b) dense glass. An example is silica gel used in chromatography and for thermal insulation. (a) Medical Applications: Hydrogels are extensively applied in medicine for purposes such as topical drug delivery, burn dress- ings, medical electrodes and breast implants. They can be used as tissue phantoms in medical physics because of their tissue-like Potential difference between anode and cathode properties. They are also used as an ultrasound gel to acoustically couple the transducer to the skin surface. More advanced uses include radiosensitive gels for dosimetry, biosensors responsive to FIGURE G.10 Operating characteristic of a gas counter: (a) propor- specific molecules, and in tissue engineering in the form of scaf- tional region of the amplitude of electrical pulses to the energy of radiation deposited in gas; (b) plateau region with the maximum gas amplification folds containing human cells to repair tissue. and the pulses not depending on the energy of radiation; (c) continuous Related Articles: Adhesives, Fricke based gel, Gel dosimetry, discharge region. Polymer gels, Ultrasound Count rate Gel dosimetry 413 General public exposure Gel dosimetry 4 (Radiotherapy) Absorbed dose measurement using nuclear mag- R2(I) = 1.22 + 0.276D R^2 = 1.000 netic resonance (MR) was first introduced by Gore et al. (1984), R2(II) = 1.28 + 0.247D R^2 = 0.999 who proposed that the changes induced by the interactions of R2(III) = 1.27 + 0.253D R^2 = 0.988 ionising radiation in the Fricke ferrous sulphate dosimeter solu- tion could be probed with MR rather than using the different light absorption properties that can be quantified by UV spectrometry. 3 Using the MR specific relaxation time parameters T1 and T2 it is possible to measure the radiation dose of the well known Fricke dosimeter. Later the Fricke solution was incorporated in a gel matrix thus realising a direct 3D dosimeter. The purpose of the gel matrix is to prohibit radiation product movement from the place of formation to surrounding regions thus saving the record 2 of 3D spatial dose distribution. The number of ferric ions being Gel I produced have been shown to be increased from approximately 15 Gel II Fe3+ ions per 100 eV to more than 100 Fe3+ ions per 100 eV. Organic Gel III additives, such as benzoic acid, enhance the Fricke dosimeter chemical yield reacting with O2. The resulting peroxides oxidise ferrous ions into ferric ions thus triggering the increase of the 1 0 2 4 6 8 dose-response sensitivity. In practice the ferric ions produced by Dose (Gy) the absorption of radiation diffuse readily through the gel or aga- rose matrix, leading to a decrease in signal intensity, and a loss of spatial information. Therefore MR imaging of the irradiated FIGURE G.12 The dose dependence of the transverse relaxation rate gel must be performed within a few hours of irradiation to avoid (R2) as a function of dose for different gel batch. (From Maryansky, M.J. et al., Phys. Med. Biol., 39, 1437, 1994.) serious degradation of the dosimetric detail. A unique and impor- tant characteristic of the gel dosimetry is that the measurement is totally non-invasive. The gel is a tissue equivalent material and A calibration is then established by plotting the estimated R2 there is the coincidence of the irradiated medium and dosimeter G for each known dose value which should fit the linear equation which can be moulded into arbitrary shapes. Therefore there is no necessity to place a probe into a phantom thus causing a perturba- tion of the radiation fluence. It is also not necessary to remove part R R 2 2 ( ) D D = D + R0 D 2 of the irradiated material to perform the measurement. The radia- D tion chemistry of the dosimetry system and the MR processes where are well understood. Another specific characteristic of the gel R0 dosimeter is that the point of measurement in gel is determined 2 is the intercept completely by the measuring system which can be programmed ΔR2/ΔD is the slope of the fitted line to the calibration data to scan the full volume to obtain a complete 3D dose distribution. The dose integration in the gel dosimeter permits measurements Figure G.12 shows values of transverse relaxation rates (R2) of dynamic and multiple beam treatments. The strong limitation for specific gels as a function of dose. The data show that the dose of the Fricke infused gel dosimetry is the post irradiation diffu- response is well fitted by a straight line and is highly reproduc- sion of the ferric ions which tends to blur the image altering the ible over a wide range of doses. An indication of the sensitivity determination of the dose distribution over time thus demanding of the gel dosimeter can be gauged by the ratio of the slope to the that the MR reading of the dosimeter take place as soon as pos- intercept. sible after irradiation. Related Article: Polymer gel dosimetry The accuracy and the sensitivity of an individual gel batch Further Readings: Appleby, A., E. A. Christman and A. depend on the preparation conditions and the chemical purity of Leghrouz. 1987. Imaging of spatial radiation dose distribution in components; therefore it is recommended to perform a separate agarose gels using magnetic resonance. Med. Phys. 14:382–384; calibration of the batch at the time of use. Several methods of gel Gambarini, G., S. Arrigoni, M. C. Cantone, N. Molho, L. Facchielli dosimeter calibration have been reported but in many of them a and A. E. Sichirollo. 1994. Dose-response curve slope improve- quantity of the gel batch is transferred to a calibration phantom ment and result reproducibility of ferrous-sulphate-doped gels and then irradiated at a series of known doses. To obtain the cali- analysed by NMR imaging. Phys. Med. Biol. 39:703–717; Gore, J. bration plot R2 = 1/T2 values must be determined for each incre- C., Y. S. Kang and R. J. Schulz, 1984. Measurements of radiation mental dose. This is achieved by measuring the signal intensity dose distributions by nuclear magnetic resonance (NMR) imag- within each flask at each echo time and fitting data by ing. Phys. Med. Biol. 41:1189–1197; Maryansky, M. J. et al. 1994. Magnetic resonance imaging of radiation dose distributions using a polymer-gel dosimeter. Phys. Med.

Biol. 39:1437–1455; Schulz, ln S (TE ) = ln S0 - R2 (D)TE R. J., A. F. deGuzman, D. B. Nguten, and J. C. Gore. 1990. Dose response curves for Fricke-infused gels as obtained by nuclear where magnetic resonance. Phys. Med. Biol. 35:1611–1622. S(TE) is the measured signal intensity at a given echo time TE S0 is the signal at TE = 0 General public exposure R2(D) is the transverse relaxation rate being function of the (Radiation Protection) General public exposures are incurred dose D by members of the public from radiation sources, excluding R2 (1/s), 64 MHz Generalised detective quantum efficiency 414 Generator, battery powered any occupational or medical exposures and the normal local geometry). The generalised MTF was developed to incorporate background radiation, but including exposures from authorised both scatter (due to the patient) and focal spot blur (resulting from sources and practices and from intervention situations. geometric unsharpness due to the finite size of the x-ray focal Also known simply as public exposures, they have defined spot). dose Limits. The generalised MTF at the detector plane at spatial frequency Related Articles: Public exposures, Dose limits ƒ is described by the equation, Further Reading: IAEA. 1996. International basic safety standards for protection against radiation and for the safety of GMTF ( f ,r,m) = ((1- r)MTFF éë(m -1) f ùû + rMTFS ( f )) radiation sources. Safety Series No. 155, International Atomic Energy Agency, Vienna, Austria. ´MTFD ( f ) Generalised detective quantum efficiency where (Diagnostic Radiology) Formed independently from the effective MTFD( f) is the detector MTF, the Fourier transform of the line detective quantum efficiency (eDQE), the generalised detective spread function (LSF) of the detector. quantum efficiency (GDQE) of a detector provides single to noise MTFS( f) is the scatter MTF and is defined as the Fourier trans- characterisation of an entire imaging system through inclusion of form of the derivative of the scatter edge spread function (ESFS). the effects of focal spot unsharpness and patient scatter to the ESFS is calculated as measure of detector DQE. The GDQE for spatial frequency ƒ in the detector plane, scat- ter fraction ρ, entrance exposure X and object magnification m is ESFS = ESFS +P - ESFP described through the equation, where ESFS+P is the edge spread function with scatter and primary GDQE ( ) GNEQ ( f ,r, X,m) radiation (with a phantom present), and ESFP the edge spread f ,r, X,m = function measured with merely primary radiation. F0 ( ) X,m MTFF[(m − 1)f] is the focal spot MTF and is the Fourier transform of the focal spot line spread function, measured using G where GNEQ(ƒ,ρ,X,m) is the generalised noise equivalent quanta a slit camera. The focal spot MTF is described in terms of the (NEQ), defined as frequency (m − 1)ƒ to translate from the detector plane to the plane of the focal spot. GMTF2 ρ is the scatter fraction, defined as GNE ( ) ( f ,r,m) Q f ,r,m = GNNPS( f ,r, X,m) S r = S + P where GMTF is the generalised modulation transfer function (see corresponding article for more details) and GNNPS is the gen- where S is scatter component of radiation reaching detector eralised normalised noise power spectrum. The GNNPS in the and P is primary component (hence (1 − ρ) is the primary frac- object plane is described by the equation tion). The scatter fraction can be estimated using the beam stop technique. GNNPS( 1 f ,r, X,m) = 2 NNPSD ( f / m,r, X ) Related Articles: Modulation transfer function (MTF), m Generalised detective quantum efficiency (GDQE), Effective detective quantum efficiency (eDQE), Detector quantum effi- where NNPSD is the normalised noise power spectrum of the ciency (DQE), Edge spread function (ESF), Line spread function detector. (LSF) To attain GDQE the GNEQ is normalised by quantum flu- Further Reading: Kyprianou, I. S. et al. 2005. Generalizing ence at the object plane, Φ0(X, m), where F X m m2 0 ( , ) = Fin ( X ) the MTF and DQE to include x-ray scatter and focal spot unsharp- . ness: Application to a new microangiographic system. Med. Phys. Φin(X) is the number of quanta measured at the detector entrance. 32(2). Related Articles: Modulation transfer function (MTF), Generalised modulation transfer function (GMTF), Effective detective quantum efficiency (eDQE), Detective quantum effi- Generator ciency (DQE), Edge spread function (ESF), Line spread function (General) See AC generator (LSF) Further Reading: Kyprianou, I. S. et al. 2005. Generalizing Generator, battery powered the MTF and DQE to include x-ray scatter and focal spot unsharp- (Diagnostic Radiology) This x-ray generator is usually used in ness: Application to a new microangiographic system. Med. Phys. low power mobile x-ray equipment. The principal diagram of a 32(2). battery-powered x-ray generator is given on Figure G.13. The DC voltage from the battery (accumulator) is converted to pulses by Generalised modulation transfer function a DC/AC converter (inverter or chopper). The pulses are usually (Diagnostic Radiology) The modulation transfer function (MTF) with frequency 0.5–2 kHz. These present AC voltage, which sup- of a detector represents its ability to resolve as a function of spa- plies the high voltage transformer and produces the x-ray tube tial frequency. Although useful to characterise the properties of anode voltage (kV). Due to the higher frequency a ferrite-core a detector, detector MTF does not characterise an entire imaging transformer can be used, what virtually transfers this generator to system (i.e. images obtained with a realistic imaging source and a type of medium (high) frequency generator. Generator, capacitor-discharge 415 Geometric efficiency DC–AC HF–HV ferrite Battery converter transformer Rectifier x-ray tube ~ 0.5–2 kHz FIGURE G.13 Block diagram of battery-powered x-ray generator. Related Articles: High voltage generator, High voltage trans- former, Medium (high) frequency generator Generator, capacitor-discharge (Diagnostic Radiology) See Capacitor-discharge generator Generator, falling load (Diagnostic Radiology) See Falling-load generator FIGURE G.14 MR image of Eurospin TO2 test object. Generator, high-frequency (Diagnostic Radiology) See High-frequency generator Geometric distortion is commonly assessed through the mea- surements of distances on images phantoms containing straight Generator kV waveform lines or arrays of points with known lengths. An example is the G (Diagnostic Radiology) See Voltage waveform TO2 Eurospin test object (Figure G.14), where four 120 mm lengths are arranged in a square. The mean of these lengths is determined and a percentage distortion can then be calculated. Generator, single-phase (Diagnostic Radiology) See Single phase generator Generators, radionuclide EXAMPLE: GENERAL GEOMETRIC DISTORTION (Nuclear Medicine) See Radionuclide generators Percentage distortion Generator(s), three-phase (Diagnostic Radiology) See Three-phase generator = (True length - Observed length) True length *100% Genetically significant dose (GSD) The percentage distortion gives an overall estimate of dis- (Radiation Protection) The genetically significant dose (GSD) tortion over the distances measured. It will not show local is an indication of the overall genetic risk. Typically the human distortions in one area of the image nor will it indicate the beings are exposed to different kind of ionising radiation type of distortion present. The percentage distortion should exposures: natural background, occupational, consumer prod- be less than 5%. ucts, environmental sources, nuclear power and medical expo- sures (diagnosis and therapy). The natural background which includes radon (the largest and most variable component), Geometric distortion cosmic rays, internal radioactivity and terrestrial radioactiv- (Ultrasound) A certain amount of geometric distortion is ity, is by far the biggest source of radiation. Man-made radia- always present in ultrasound imaging. This is inherent in the tion includes medical exposures and consumer products. The technique as the time for an echo to return is assumed to be pro- genetically significant dose can be estimated for populations, portional to distance. At the same time, the presence of echoes groups of population (as example occupational workers, who is dependent on mismatches in speed of sound. In the body, should pay particular attention) and also deriving from particu- an average speed of sound of 1540 m/s is assumed. Fat on the lar items to a population (as, e.g. was done for luminous wrist other hand has a speed of sound which is around 1430 m/s, and watches, etc.). a fat layer will thus be represented as too thin in the image. Hyperlinks: https//www .NCRP .org; https//www .IAEA .org Normally, this effect is small (<5%), but differences in speed of sound can also lead to degraded image quality: a focussed beam Geometric distortion will defocus through a fat layer, leading to less sharp ultrasound (Magnetic Resonance) Geometric distortion is the deviation of images. points in an image from their true position in an object. This arte- fact appears as either displacement of a point or improper scaling Geometric efficiency and is primarily caused by a non-uniform main magnetic field or (Diagnostic Radiology) In CT scanning, geometric efficiency non-linear magnetic field gradients. refers to the geometric dose efficiency of the detector array. In Geometric error 416 Geometric field size general terms, it is the percentage of photons exiting the patient Further Reading: Medical electrical equipment Part 2–44: that fall on the active detector area. Geometric efficiency is Particular requirements for the safety of x-ray equipment for com- reduced by inactive areas within the detector array, such as the puted tomography, EN 60601-2-44:2001 + Amendment 1 (2003). septa between detector elements, or if the x-ray beam is greater than the active detector area. Geometric error More specifically, the term geometric efficiency (GE) is used (Magnetic Resonance) Geometric distortion can be described as the when referring to the geometric dose efficiency along the the positive distortion (geometric error) in the x, y and z planes of z-axis (scan axis), which is affected by the x-ray beam extend- the image. Geometric error is a useful way of quantifying geomet- ing beyond the active detector length (see Overbeaming). A more ric distortion and can be used to monitor the systems geometric complete term for this geometric efficiency is therefore ‘z-axis distortion over time and also compare it with other systems. For a geometric efficiency’. This definition does not take into account modern MR system a maximum geometric error on the order of inefficiencies caused by septa between detector banks. The defi- 1–2 mm in any direction would be expected. nition of z-axis GE in CT is given in the following equation and The geometric error can be measured simply by making length shown diagrammatically in Figure G.15: measurements from readily identified locations on an image in the three different planes and comparing them with the known values for those lengths. Area of dose profilewithin total imagedslice width A GE = = i Related Article: Geometric distortion Totalarea underdose profile Atot Geometric field separation On multislice CT scanners, the z-axis geometric efficiency varies (Radiotherapy) Sometimes the area to be treated cannot be with the z-axis beam collimation, greater collimations generally covered by one field. If this is the case, then two (or more) have a higher geometric efficiency. Figure G.16 shows an example fields may be applied. Because radiation fields diverge it is of the variation in z-axis GE with collimation on a 16-slice CT important to know the point where the geometric edges meet scanner. or overlap in order to provide an effective treatment and to A more complete definition of z-axis GE in CT can be found in avoid over-radiation or under-radiation. Calculating the sepa- European Standard EN 60601-2-44. ration of fields can be done with respect to the geometric edges Related Articles: Dose profile, Overbeaming, Multislice CT G of the beam (see also Geometric field size) or the central axis scanner of the beams. Figure G.17 shows two beams set to overlap at a depth d below the surface of the patient. The overlap point can of course be set to any convenient depth by altering the central axis separation or the geometric field edge separation. Related Article: Geometric field size Dose profile Geometric field size (Radiotherapy) The shape and size of the radiation field is affected i z-axis by the size of the source and the effect of scatter. In the case of a Tot linear accelerator the effective source size would be the electron spot which is on the order of a few millimetres and in the case of Detectors cobalt it is around 15 or 20 mm. Figure G.18 shows the result for a typical source. The geometric

field size is generally defined as the FIGURE G.15 z-axis geometric efficiency: percentage of dose profile distance between the 50% isodose lines at opposite sides of the utilised for imaging. (Courtesy of ImPACT, UK, www .impactscan .org) field, the maximum dose being taken as 100% (see Figure G.18). Z-axis geometric Collimation efficiency (%) 5 mm (4 × 1.25) 67 10 mm (8 × 1.25) 83 20 mm (16 × 1.25) 97 5 mm 10 mm 20 mm Z-axis FIGURE G.16 Effect of z-axis x-ray beam collimation on geometric efficiency. Geometric unsharpness 417 Germanium detector Central axis Further Readings: Bental, G. C. 1996. Radiation Therapy separation Planning, 2nd edn., McGraw-Hill, New York, p. 48; Bomford, C. K. and I. H. Kunkler. 2002. Walter and Miller Textbook of Radiotherapy, 6th edn., Churchill Livingstone, Oxford, UK, pp. 172, 200; Mayles, P., A. Nahum and J. C. Rosenwald. 2007. Handbook of Radiotherapy Physics: Theory and Practice, Taylor & Francis, Boca Raton, FL, pp. 458–460. Patient surface Geometric unsharpness (Diagnostic Radiology) Geometric unsharpness is a term used in the past to refer to what is better described as focal spot blur- ring in x-ray imaging. The amount of focal spot blurring is related to the geometry and relative positions of the focal spot, receptor, D and object within the body as illustrated on Figure G.19. Here the Geometric object (a hole in an object) is projected over the receptor/film (the beam white spot) together with some blur (the dark spot). The higher is overlap point the magnification (the object is far away from the film) the larger Geometric beam edge are both the white and blue spots, but their superposition leads to separation domination of the blur over the object image. As the object in the patient’s body is located away from the FIGURE G.17 Geometric field separation. receptor (increased geometric magnification) the blurring because of the finite size of the focal spot is increased relative to the size of the object’s image. This increases the blurring and reduces vis- ibility of detail. Source The amount of blurring relative to the size of an object is related to the size of the focal spot and the location of the object Collimators relative to the receptor and focal spot (along the ‘s’ scale) – shown in Figure G.20. G Blurring from the x-ray image receptor also has the same geo- metric dependence as focal spot blurring (Figure G.21). Penumbra As geometric magnification is increased, the size of an object ‘as seen by the receptor’ is larger. Therefore the amount of blur from the receptor and relative to the size of an object is decreased and visibility of detail is improved. This is the reason to use geo- metric magnification in procedures like mammography when high visibility of detail (small calcifications is required). The amount of blurring relative to the size of an object is related to the actual blur value within the receptor and the loca- tion of the object relative to the receptor and focal spot (along the Geometric 100% field ‘s’ scale) as shown on Figure G.22. Germanium detector Percentage (Radiation Protection) Germanium (Ge) is a semiconductor 50% depth dose with atomic number Z = 32. The energy needed to create an FIGURE G.18 Source size and scatter have an effect on the size of the penumbra of a radiation field. Magnification The field size is defined for the specified source – surface distance at the specified depth. Geometric field size is usually referred to as radiation field size or radiation beam field size. The source size together with the collimator transmission and scatter affects the width of penumbra region. The larger the source, the higher the portion of the source, which is shielded by the collimator, and the broader the resultant penumbra. Cobalt beams therefore will have a broader penumbra than linear accel- Focal spot blur size erators because they have a larger effective source size. Related Articles: Penumbra, Beam limiting device, FIGURE G.19 Focal spot blurring – influence of the image magnifica- Collimator, Collimation tion on the blur. (Courtesy of Sprawls Foundation, www .sprawls .org) Germanium detector 418 G ermanium detector Focal spot Focal spot location F (mm) 1.0 1.0 0.9 0.9 0.8 Blur (mm) = F (mm) × s 0.8 0.7 Anatomical 0.7 object 0.6 0.5 0.6 0.4 0.5 0.3 Anatomical 0.4 Blur (mm) = R (mm) [1–s] object 0.2 0.3 0.1 0.2 0.0 0.1 R (mm) 0.0 Receptor equivalent Receptor location blur FIGURE G.20 Relation between the focal spot, object location (mag- FIGURE G.22 Relation between the detector unsharpness, object loca- nification) and blur size. (Courtesy of Sprawls Foundation, www .sprawls tion (magnification), and blur size. (Courtesy of Sprawls Foundation, .org) www .sprawls .org) G Magnification 1 0.5 FWHM Receptor and display blur size Energy FIGURE G.21 Detector blurring example. (Courtesy of Sprawls Foundation, www .sprawls .org) FIGURE G.23 Example of energy spectrum recorded with a Ge(Li) detector. electron-hole pair is about 3 eV, much less than the 35 eV required in gas detectors. The density of Ge is much greater than the den- electronic noise. The relative contribution of these factors depends sity of gas used in gas-based detectors. Thus, in Ge detectors on the energy of the x-rays or gamma radiation and on the volume about 10 times more ion pairs are produced with greater prob- of the detector. Therefore the energy resolution is specified at par- ability (efficiency) than in a gas detector for a given x- or gamma- ticular photon energies, e.g. 5.9 keV (Fe-55), 122 keV (Co-57), 662 ray. The energy resolution of Ge detectors is also better than gas keV (Cs-137), 1.33 MeV (Co-60). Typical FWHM values for small detectors. The atomic number of Ge is greater than that of silicon volume Ge detectors are 150–250 eV at 5.9 keV and 400–600 eV (Z = 14) thus Ge detectors have a larger cross-section for photon at 122 keV. For larger volume detectors typical values are 0.8–1.2 interactions and better detection efficiency than silicon detectors. keV at 122 keV and 1.7–2.3 keV at 1.33 MeV. There are two types of Ge detectors: those doped with lithium The germanium detector has large thermal noise at room (GeLi); and high-purity Ge (HPGe). The ionisation in the detector temperature which causes a high background reading. This back- results in a small electrical current and is converted to a voltage ground signal decreases as the temperature of the detector is pulse. The height of the pulse is proportional to the energy of the reduced. Liquid nitrogen, at a temperature of 77 K, is used to cool radiation absorbed by the detector. the detector. Lithium drifted germanium Ge(Li) must be kept at The energy resolution of a semiconductor detector is defined this temperature to prevent the lithium drifting out. HPGe detec- by the full width at half maximum (FWHM) of the photopeak tors, however, are stable at room temperature. (Figure G.23). Its value is expressed in keV and not as a percent- Further Readings: Dendy, P. P. and B. Heaton. 1999. Physics age of the photon energy. The energy resolution depends on the for Diagnostic Radiology, IOP Publishing, Philadelphia, PA; statistical fluctuation in the number of created electron-hole pairs, Knoll, G. F. 2000. Radiation Detection and Measurement, John variation in the charge collection efficiency with detector size and Wiley & Sons, Inc., New York. Relative object location (‘s’ scale) Relative intensity Relative object location (‘s’ scale) Geometric (g)-factor 419 Gibb’s ringing Geometric (g)-factor differences in alternate lines in k-space. The ghost produced is (Magnetic Resonance) In parallel acquisition techniques (PAT), displaced half the field of view away from the original image. signals are combined from individual coil elements to enable Quadrature Ghost: RF quadrature artefacts are caused by synthesis of some of the phase-encoded signals, reducing over- a disturbance in the two detector channels of the quadrature all scan time. Signal to noise ratio (SNR) in parallel imaging is detector. DC offset of one output of one of the amplifiers will reduced by the square root of the acquisition reduction time (or produce a bright point in the centre of the image. A higher gain acceleration factor), but also reduced by a parameter related to of one detector will result in a ghosting of objects diagonally the geometry of the coils relative to the objects being imaged. in the image. This parameter is known as the geometric factor or g-factor. The Abbreviations: DC = Direct current, EPI = Echo planar imag- g-factor is spatially varying causing a non-uniform noise across ing, RF = Radiofrequency and SNR = Signal to noise ratio. images acquired with parallel imaging techniques. Related Articles: Motion artefacts, N/2 artefact, Quadrature Related Articles: Parallel acquisition technique (PAT), Phased artefact array coil, SNR Further Readings: Pruessmann, K. P., Weiger, M., Ghosting Scheidegger, M. B. and Boesiger, P. 1999. SENSE: Sensitivity (Magnetic Resonance) Ghosting is the term used to describe a encoding for fast MRI. Magn. Reson. Med. 42(5):952–62; low intensity displacement of MR signal which can manifest itself Robson, P. M., Grant, A. K., Madhuranthakam, A. J., Lattanzi, either as a series of distinct low intensity copies (ghosts) of the R., Sodickson, D. K. and McKenzie, C. A. 2008. Comprehensive image, or as a column of smeared signal. quantification of signal-to-noise ratio and g-factor for image- The most common cause of ghosting is motion; this can be based and k-space-based parallel imaging reconstructions. Magn. patient motion, physiological motion or blood flow. Ghosting Reson. Med. 60(4):895–907. mainly propagates along the phase encoding direction since sig- nal sampling and spatial encoding are performed over many TRs Ghost artefact and hundreds of milliseconds in the phase encoding direction, (Magnetic Resonance) A ghost artefact is a low intensity image of compared to a few milliseconds in the frequency encoding direc- an object which appears in a different location from the primary tion (Figure G.24a and b). signal area. When ghosting occurs there is a SNR reduction in the Clinically, techniques such as breath hold imaging and cardiac primary signal area. Ghosting can potentially lead to important gated imaging are often applied to reduce the effects of ghosting clinical information being obscured by the ghost signal. from physiological motion. G Motion Ghost: This type of ghost is caused by movement of Related Article: Cardiac gating the patient, either physiological motion such as breathing and bowel motion, or patient movement such as small twitches. The ghost appears along the phase-encoding direction. Gibb’s artefact Motion ghost artefacts can be reduced by respiratory and cardiac triggering, the (Magnetic Resonance) See Gibb’s ringing use of breath holding pulse sequences, or presaturation pulses, depending on their origin. Gibb’s ringing Flow Ghost: Flow artefacts are another type of motion arte- (Magnetic Resonance) Gibb’s ringing artefacts (also known as fact caused by the flow of blood or cerebrospinal fluid within the truncation artefacts) occur near sharp boundaries and high- body. Flow compensation can be used to reduce this artefact. contrast transitions in MR images. The artefacts appear as alter- Nyquist Ghost: The N/2 or Nyquist artefact only occurs when nate bands of high and low intensity signal, parallel to interfaces using an EPI pulse sequence. The artefact is caused by slight between tissue or phantom edges. (a) (b) FIGURE G.24 (a) Ghosting along the phase encoding direction due to eye movement. (b) Low intensity whole head ghosts (along the phase encoding direction). Glass envelope (of an x-ray tube) 420 Glass envelope (of an x-ray tube) Clinically, there are two main anatomical locations where voxels in this row represent regions outside the test object, they Gibb’s ringing occurs: should have an intensity value of zero, whilst the central voxels (representing the test object itself) should be of constant, non- 1. At the edge of the brain zero intensity. The off-on-off waveform representing this inten- 2. At the interface between CSF and the spinal cord sity variation is known as a ‘top-hat’ function (see Figure G.25). According to Fourier theory, the ‘top-hat’ of voxel intensi- The cause of Gibb’s ringing is insufficient sampling of the high ties can be approximated by summing a series of sinusoids. In frequencies inherent at sharp edges. Figure G.26, it can be seen the more sinusoids are summed, the Fourier Theory: To understand the phenomenon further, it is better the approximation. Yet, even when summing

large num- useful to consider Fourier theory, which states that any periodic bers of sines, the resultant waveform still deviates slightly from waveform can be constructed from a sum of sinusoids of varying the ideal: the sharp edges of the ‘top-hat’ are accompanied by an amplitude, frequency and phase. unwanted oscillation of signal – this is Gibb’s ringing. In MRI, images are created via such a summation, where the In clinical MRI, Gibb’s ringing is often exacerbated when number of frequencies present in the final image is determined by the number of phase encoding steps is reduced to reduce image the number of phase and frequency encoding steps. For a simple acquisition time. If the number of phase encoding steps is halved example, let us consider a row of pixels through an MR image from 256 to 128, the acquisition time will also be halved, but so of a uniform test object (see Figure G.25). As the first (and last) will the number of frequencies present in the final MR image. For this reason, reducing the number of phase encoding steps increases the amount of oscillatory signal (Gibb’s ringing) arte- fact. Figure G.27a and b demonstrate this in practice. Gibb’s artefact can be avoided by increasing the number of Pixel intensity phase encoding steps or changing the resolution of the image. Glass envelope (of an x-ray tube) (Diagnostic Radiology) The glass envelope provides the support of the anode and the cathode assemblies. It also keeps the deep G FIGURE G.25 An idealised MR image of a uniform phantom, with a vacuum in the x-ray tube (min 10−6 mbar) and the high voltage corresponding ‘top-hat’ plot of intensity variation along a given row of insulation between the anode and cathode. The glass should have pixels. low x-ray absorption and high thermal resistance. 1 sine 3 sines 5 sines 10 sines 50 sines 256 sines FIGURE G.26 Top-hat functions composed from varying numbers of sine waves. The initial sine waves added to the sum are of low frequency, but higher frequency sines add finer detail. (a) (b) FIGURE G.27 (a) Plot of pixel intensity variation at an air/phantom interface, as depicted by 256 phase encoding steps. (b) Plot of pixel intensity variation at an air/phantom interface, as depicted by 128 phase encoding steps. Gloves, lead 421 G MR (gradient motion rephasing) 100 200 300 400 FIGURE G.28 X-ray tube with rotating anode (medium power diag- Temperature (°C) nostic tube) – zoomed view. Note the cracked glass against the anode. (Courtesy of EMERALD project, www .emerald2 .eu) FIGURE G.29 Example of a thermoluminescent glow curve. When multiple heavy exposures are made the anode tem- of the glow curve, e.g. using an appropriate software is applied perature is so high, that it glows (white hot). This heat trans- to determine the radiation dose in routine personal dosimetry fers to the near glass envelope. The heating and later cooling practice. produces thermal stress, which leads to micro-cracks, which Related Article: Thermoluminescent dosimeter (TLD) decreases the vacuum and produces arcing inside the x-ray Further Readings: Horowitz, Y. 2006. Microdosimetric tube (Figure G.28). Response of Physical and Biological Systems to Low and High- When producing the x-ray tube the glass envelope is first vacu- LET Radiations: Theory and Applications to Dosimetry, Elsevier, umed and then sealed. Even so with the time the glass and the Amsterdam, the Netherlands, pp. 280–282; Stabin, M. G. 2008. G other parts in the envelope emit ions (cold emission). Special care Radiation Protection and Dosimetry: An Introduction to Health is taken for diminishing of the cold emission – like polishing and Physics, Springer Science + BusinessMedia, LLC, New York, pp. degassing the glass and the metal electrode assemblies. However 158–159. small, this emission exists and when the tube is not used for a long time it can lead to arcing and damage the tube. Due to this reason x-ray tubes which have not been used for a long time must Glycolysis be ‘degassed’ – warmed slowly with low-power exposures before (Nuclear Medicine) Glycolysis is the metabolic pathway from use (this will be explained later). glucose to pyruvate with a net production of two molecules of Often the glass is thinned at the exit of the x-rays (exit win- ATP per molecule of glucose. Pyruvate can then be converted to dow) to minimise the absorption of useful radiation. Exactly lactate or enter the tricarboxylic acid cycle. Complete oxidation of at this place the metalisation of the glass is quite substantial glucose through glycolysis and the tricarboxylic acid cycle yields and increases with the age of the x-ray tube (see the article on a net production of 38 molecules of ATP, 6 molecules of CO2 and Stationary anode). 6 molecules of H2O per molecule of glucose. The disadvantages of the glass envelope are overcome in the metal envelope x-ray tubes. Glycolysis targeting The anode support (the shank at which the rotor with the rotat- (Nuclear Medicine) This term refers to radiopharmaceuticals that ing anode is coupled by the bearings) emerges from the glass target the cellular glucose metabolism. The most common glu- envelope and is connected to the positive end of the high volt- cose targeting radiopharmaceutical is 18F-FDG (2-deoxy-2-[18F] age. Special air-tight and heat-resistant seal is used to fix the glass fluoro-d-glucose) which is used for clinical PET imaging. FDG is envelope to the anode shank. transferred into the cell by the gluc-1 transporter which is pres- Related Articles: Filament circuit, Filament heating, X-ray ent in many cell types but overexpressed in cancer cells. Once tube, Anode, Focal spot inside the cell, the FDG undergoes phosphorylation by hexoki- Hyperlinks: EMERALD (DR module), www .emerald2 .eu nase which traps the FDG inside the cell. A PET imaging study will show the FDG distribution (or glucose metabolism) within Gloves, lead the patient. This makes FDG-PET an excellent tool to detect, (Radiation Protection) See Lead gloves monitor and classify tumours. Related Articles: Tracer kinetic modelling, Receptor target- Glow curve ing, Antigen targeting, DNA targeting, Neuroreceptor targeting, (Radiation Protection) The glow curve is the emission spectra Apoptosis targeting, Hypoxia targeting (Figure G.29) registered during heating of thermoluminescent Further Reading: Imam, S. K. 2005. Molecular nuclear material that was previously irradiated. imaging: The radiopharmaceuticals (review). Cancer Biother. The glow curves of different luminescent material have well Radiopharm. 20(2):163–172. known shapes. The parameters are peak energy and full width at half maximum as well as its surface. As the light output is lin- GMR (gradient motion rephasing) ear dependent on the radiation energy and dose, the evaluation (Magnetic Resonance) See Gradient motion rephasing (GMR) Relative intensity Golay codes 422 Gradient coils Golay codes Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in (Ultrasound) A Golay coded transmission sequence is used in Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American coded excitation schemes. The Swiss-born mathematician Marcel Pharmacists Association, Washington, DC; Zolle, I., ed. 2007. Golay, discovered that certain pairs of binary sequences have Technetium-99 m Pharmaceuticals – Preparation and Quality important correlation properties. Such pairs have equal length Control in Nuclear Medicine, Springer, Heidelberg, Germany. and the sum of their autocorrelation functions is zero except at zero lag. Examples are for instance [−1, +1], [+1, +1] and [−1, Gradient and spin echo (GRASE) +1, −1, −1], [−1, +1, +1, +1]. Thus, range lobes are theoretically (Magnetic Resonance) This is a hybrid pulse sequence that eliminated. Golay coded transmission schemes seem to be imple- combines multiple gradient and fast spin echo imaging. The mented successfully in different ways. One approach employs pulse sequence consists of a train of refocusing RF pulses and transmission of a sine burst where +1 in this sequence corre- alternating readout gradients. The refocusing RF pulses lead to sponds to a single cycle sine wave, and −1 to an inverted cycle (i.e. a train of spin echoes, whereas the alternating gradients create 180° out of phase). Another is to sequentially excite the transducer multiple gradient echoes. The GRASE pulse sequence is illus- with a positive pulse, or a negative one, respectively, according to trated in Figure G.30, with three readout lobes between each this scheme. Sufficient time is allowed between excitations to let of the 180° refocusing pulses. GRASE images can be acquired the impulse response ring down. At least one manufacturer has with higher spatial resolution than those obtained from the fast employed Golay codes to improve SNR and penetration depth. spin echo sequence since more echoes per unit time can be Apparently the Golay coding is only used for larger depths, as obtained using alternating readout gradients. The SAR is also the focusing at shallower regions distorts the coding to a larger lower than for RARE because fewer RF pulses are used, allow- extent. A drawback to using Golay codes is the need for multiple ing more slices per TR, especially at high field strength (e.g. transmits, thereby reducing frame rate. 3.0 T). Drawbacks of this pulse sequence include sensitivity to Further Readings: Chiao, R. Y. and L. J. Thomas. 2000. eddy currents (as known from EPI imaging), leading to phase Method and apparatus for ultrasonic beamforming using Golay- errors. Phase corrections are necessary to remove inconsisten- coded excitation. US patent 5,984,869; Leavens, C. et al. 2007. cies between echoes originating from the alternating polarity Golay pulse encoding for microbubble contrast imaging in readout. ultrasound. IEEE Trans. Ultrason. Ferroelectr. Freq. Control Related Articles: Eddy currents, EPI (echo planar imaging), 54:2082–2090; Nowicki, A. et al. 2007. Influence of the ultra- G Fast spin echo, Specific (energy) absorption rate, SAR sound transducer bandwidth on selection of the complementary Further Reading: Bernstein, M. A., K. F. King and X. J. Zhou. Golay bit code length. Ultrasonics 47:64–73. 2004. Handbook of MRI Pulse Sequences, Elsevier Academic Press, San Diego, CA. Gonad shielding (Radiation Protection) It is a good practice to protect the gonads Gradient coils of young people during x-ray examinations. (Magnetic Resonance) Gradient coils generate the magnetic field For male patients there are specially shaped scrotal protec- gradient in the x, y and z direction (see related article) when cur- tions (gonad shields), in different size, usually at least four sizes, rent is fed through them. The gradient coils are designed in a with a protection in Lead equivalent thickness of 2 mm. way to create spatial magnetic field gradient as linear as possible. For female patients there are specially shaped ovary protec- Furthermore, the gradient fields have to be switched on and off as tions in different sizes, usually five sizes, with a protection in fast as possible during the execution of a pulse sequence. Lead equivalent thickness of 1 mm. The hardware required to generate the gradient fields is a set For small female patients, special attention should be paid to of coils (electromagnets) mounted on a cylindrical former. This the choice and positioning of the protection, taking into account is most easily illustrated for the gradient coil generating the gra- the anatomy and the aim of the x-ray investigation. In fact a wrong dient field in the z-direction (the direction along the tunnel of a choice or positioning might cover important anatomical struc- cylindrical magnet with a horizontal B0-field). In Figure G.31a tures and invalidate the investigation. an example of such a gradient coil consisting of two separate Further Reading: IAEA. 1996. International basic safety standard for protection against ionising radiation and for the safety of radiation sources. Safety Series No. 115, International 90° 180° 180° 180° Atomic Energy Agency, Vienna, Austria. rf Good manufacturer practice (Nuclear Medicine) Good manufacturer practice (GMP) is concerned with ensuring good quality of the production and G preparation of pharmaceuticals, control and management of man- ufacturing and quality control testing of pharmaceutical products. A principal part of GMP is the recording and documentation of G all activities during the production of the pharmaceutical enabling traceability. Abbreviation: GMP = Good manufacturer practice. G Related Articles: Quality control, Radionuclide purity, Radiochemical purity Further Readings: European Pharmacopeia, 2013. European AD Directorate for the Quality of Medicines (EDQM), Council of Europe [http: / /www .edqm .eu /s ite /H omepa ge - 62 8 .htm l]; FIGURE G.30 The GRASE pulse sequence. Gradient echo (GE) 423 G radient echo (GE) B0 [T] B0 [T] Gz [T] Gy [T] B0 + Gz  z [T] z [m] (a) (b) FIGURE

G.31 (a) Gradient coil used to generate a linear gradient field along the direction of the B0-field in a typical cylindrical MRI magnet. (b) Gradient coil configuration used to generate linear gradient fields in the x- and y-direction. windings in opposite directions (Helmholtz- or Maxwell-pair) is shown. When current (generated by the gradient amplifier) runs RF through these windings it results in gradient fields that either adds to or subtracts from the B0-field. The gradient field is linear and Slice becomes stronger with increasing distance from the midpoint. Phase The midpoint where the gradient field is zero is typically located at the isocentre of the magnet. Readout For the other two directions (x and y) a different coil configu- Signal ration is needed. It is based on a design invented by mathemati- TE cian Marcel J E Golay in 1958. Hence it is denoted Golay coil, G a type of saddle coil (Figure G.31b). The coil configurations for gradient fields in the x- and y-direction are similar, only rotated FIGURE G.32 Timing diagram for a basic gradient echo pulse sequence. 90° relative to each other. The rather loud noise heard from an MRI scanner when it operates is generated by the force that the gradient coils (located area under the two gradient forms are equal. This corresponds within a very large B0-field) experience as the current feeding to the centre of the gradient echo that is formed and is denoted them. This is much the same way a normal loudspeaker works, the echo time (TE). only with a much stronger permanent magnet and a much stronger The image contrast for a gradient echo pulse imaging sequence current (typically several hundreds of Amperes) running through depends not only on the echo time and the repetition time (TR) the windings. In order to prevent most of the movement of the but also on the excitation flip angle (α). Usually, a gradient echo coils they are moulded into an epoxy resin. Still the vibrations are image is measured by fast repetition of the sequence to establish a quite noticeable and the noise is sometimes a rather severe prob- steady state of the transverse magnetisation, which is an equilib- lem in the MRI environment, especially with high field strength rium between the small flip angle excitation and the T1-relaxation magnets and high power gradient systems. Therefore, ear protec- process. In general, smaller flip angles results in more T2*- tion is normally required for the patients. weighted images and larger flip angles result in more T1-weighted Related Articles: Eddy current, Gradient field, Isocentre, images. Specifically the following thumbs of rule apply to obtain Pulse sequence different contrasts: Further Reading: McRobbie, D. W., E. A. Moore, M. J. Graves and M. R. Prince. 2003. MRI from Picture to Proton, Cambridge T1-weighted: Large flip angle, short TR and short TE University Press, Cambridge, UK. T * 2 -weighted: Small flip angle, long TR and long TE PD-weighted: Small flip angle, long TR and short TE Gradient echo (GE) (Magnetic Resonance) A gradient echo is a signal generated There is also a relationship between TR, α and T1 of the tissue, from free induction decay (FID) by means of a bipolar switched such that there is a flip angle that will result in maximum signal magnetic gradient. A gradient echo sequence (Figure G.32) for a particular TR and T1. It is denoted the ‘Ernst-angle’. consists of slice selective excitation pulse to create transverse The lack of a refocusing (180°) pulse, which would be used magnetisation in a slice of the object. The transverse magnetisa- to obtain a spin-echo, means that spins will lose coherence in tion will decay with time (free induction decay) according to the presence of any static magnetic field inhomogeneity. This the tissue relaxation time T2*. At any time while there is still a causes severe signal loss near, for example air cavities or metal detectable signal, a gradient echo can be obtained by applying implants, but can also be very useful in, for example functional a bipolar gradient pulse in the readout direction. When the first magnetic resonance imaging applications where differences in gradient pulse in the readout direction (here negative) is applied, magnetic susceptibility actually results in the desired signal spins will quickly get out of phase and no signal can be detected. contrast. When the second pulse is applied in the opposite direction the Related Articles: Fast low angle shot, Fast field echo, Flip signal returns again and is at its maximum exactly when the angle, T1-weighted, T2-weighted y [m] B0 + Gy  y [T] Gradient field 424 Gradient linearity Gradient field The same principle is valid for the other two directions that (Magnetic Resonance) Magnetic field gradient (∂Bx/∂x, ∂By /∂y are used for spatial encoding of an MR-image (denoted phase and and ∂Bz/∂z) in three orthogonal directions (x, y and z) are used to frequency encoding), only that the gradient fields are applied at obtain spatial localisation of the signal that is measured in MRI. a different time during the execution of a pulse sequence. The The three directions correspond to the physical coordinate system frequency encoding gradient is applied during readout of the of the magnet, with z oriented along the tunnel (and along B0) for MR-signal and the phase encoding gradient is applied at some a cylindrical magnet with a horizontal B0-field. time between excitation (slice selection) and readout. The gradient fields are switched on and off during the execu- The gradient fields are switched on and off (see Figure G.34) tion of a pulse sequence. The gradient fields add to the static while still avoiding a rate of change (dB/dt) large enough to induce magnetic field (B0) resulting in a linear variation of the magnetic currents that can cause nerve stimulation. The rise times are on field in each direction. If for example a gradient in the z-direction the order of a few 100 μs and amplitudes are typically between 5 (Gz) is switched on, the total magnetic field (B) that a hydro- and 40 mT/m. The gradient fields thus are time varying magnetic gen nucleus (proton) located at position z experiences can be fields with frequencies on the order of 1 kHz, which decrease in expressed as amplitude with increasing distance from the gradient coil. Related Articles: Gradient coils, Isocentre, Pulse sequence B(z) = B0 + Gz × z éëTùû Further Reading: McRobbie, D. W., E. A. Moore, M. J. Graves and M. R. Prince. 2003. MRI from Picture to Proton, Cambridge University Press, Cambridge, UK. The Larmor frequency ( f) at that point is given by Gradient linearity f (z) - = g × (2p) 1 (B0 + Gz × z) éëHzùû (Magnetic Resonance) MR systems contain three gradients for spatial encoding in the x-direction, y-direction and z-direction. It is important that each gradient is linear in order to obtain accu- where γ · (2π)−1 ≈ 42.6 MHz/T. rate spatial localisation. Gradient linearity is defined as the dif- The gradient field is usually zero at the isocentre (x, y and z ference between the measured value of the field gradient and the = 0) of the magnet, so that the gradient field adds to or subtracts ideal measure. The non-linearity is expected to be approximately from B0 in a linear manner as the distance from the isocentre 1%–2% over a 50 cm DSV. Gradient linearity tends to decrease G increases. This results in a linearly varying Larmor frequency towards the edge of the volume being imaged (Figure G.35). This along the direction of the applied gradient field. Figure G.33 illustrates a gradient field that is applied along the direction of slice selection while an RF-pulse is transmitted. G Depending on the frequency and bandwidth of the RF-pulse the hydrogen nuclei within a slice with a specific thickness at a spe- cific location along the slice selection direction can be excited. Note that the slice selection direction can be defined freely by Amplitude the operator and the gradient field that is needed can be obtained by combining gradient fields applied in the x, y and z-directions simultaneously. Rise time Time FIGURE G.34 The typical waveform of a gradient that is ramped to specific amplitude during its rise time. B0 [T] f = y  (2π)–1 (B0 + Gslice  d) [Hz] Ideal gradient Actual gradient B0 + Gslice  d [T] Position B0 d = 0 d [m] FIGURE G.33 An RF-pulse is transmitted while a slice encoding gradi- ent is applied. Only hydrogen nuclei with a Larmor frequency equal to the RF-pulse frequency are excited, i.e. those within a slice at a specific distance (d) from the isocentre (d = 0). FIGURE G.35 Gradient non-linearity. Gradient Gradient motion rephasing (GMR) 425 Gradient spoiling results in the image being geometrically distorted at the edges of G G the object. To remove some of the problems of gradient non-linearity, –G t some manufacturers have used computer algorithms to unwarp the image, taking into account the known distortions of the real- world non-linear gradients. (a) Abbreviation: DSV = Diameter of a spherical volume. Related Article: Gradient Gradient motion rephasing (GMR) (Magnetic Resonance) In the presence of a magnetic field gradient Gslice t G, the phase shift ΔΦ (in the rotating frame of reference) expe- rienced by a spin isochromat as a function of time t is given by T DF(T ) = gòG (t )r (t )dt 0 where γ is the gyromagnetic ratio Gread t r is the position The phase development of a spin is influenced by the gradient (b) timing as well as the position of the spin at a given time, and a spin is referred to as being rephased when the net phase shift has returned to zero, i.e. ΔΦ = 0. Therefore, static spins are obviously FIGURE G.36 (a) Three-lobe gradient waveform resulting in nulling of rephased at time T if the zeroth and first gradient moments. (b) Slice-selection and frequency- encoding gradient waveforms for first-order motion compensation in a G gradient-echo sequence. T òG (t )dt = 0 0 of higher-order gradient moments tends to become increasingly complicated and time consuming to execute. The position of an arbitrarily moving spin isochromat is given Figure G.36a illustrates a general example of a three-lobe gra- by the sum of contributions from motion components of differ- dient waveform that will result in nulling of the zeroth as well ent order, i.e. initial position (zeroth order), velocity (first order), as the first gradient moment, i.e. static spins and spins with con- acceleration (second order), jerk (third order), etc. When the posi- stant velocity will be rephased. However, a number of alternative tion is described in terms of its Taylor series, i.e. gradient schemes can be designed to accomplish the same result. Figure G.36b shows examples of slice-selection and frequency- 2 r (t ) at dnr tn = r0 + vt + + + + encoding (readout) gradient waveforms for first-order motion 2 dtn n! compensation in a normal gradient-echo-type pulse sequence. The net phase shift observed at the centre of the readout period the phase shift can be expressed as (arrow) is independent of zeroth and first order motion compo- nents, i.e. static spins and spins moving at a constant velocity will é T T T be rephased. Any phase shifts resulting from higher-order motion ( ) a DF T = g êr0òG (t )dt + vòtG (t )dt + òt2G (t )dt components along the slice-selection or frequency-encoding ê 2 directions will, however, remain. ë 0 0 0 Related Articles: Flow compensation, Pulse sequence, T dn r 1 ù + + dtn òt nG (t )dt + ú Gradient field n! ú Further Reading: Haacke, E. M., R. W. Brown, M. R. 0 û Thompson and R. Venkatesan. 1999. Magnetic Resonance From the given sum, it can be seen that the nth integral expres- Imaging: Physical Principles and Sequence Design, John Wiley sion corresponds to the nth gradient moment with respect to time. & Sons, New York. The net phase shift observed at an echo in MRI can thus be made independent of the nth order motion if the nth gradient moment Gradient spoiling is zero at that time. Zeroth moment nulling corresponds to the (Magnetic Resonance) Gradient spoiling is used to

destroy the common rephasing of static spins, employed in all standard MRI coherence of an MR-signal by adding a strong uncompensated pulse sequences. gradient in a pulse sequence after signal sampling. As a conse- Gradient-design strategies aimed at nulling certain higher gra- quence of this spoiler gradient the spins at different locations dient moments are often referred to as gradient motion rephas- will experience a variety of magnetic field strengths and thus will ing, gradient moment nulling or, simply, flow compensation. precess at different frequencies. Therefore the spins will quickly Gradient schemes for first-order motion (velocity) compensation become dephased and the net magnetisation in the transverse are fairly simple in their design (Figure G.36), but the nulling direction will be reduced. Grand-daughter radionucleus 426 G rating interferometry Gradient spoiling is not a very efficient way to destroy the Graphite is an allotrope of carbon; it consists of layers of car- transverse magnetisation. In particular, spins close to the isocen- bon atoms arranged in a hexagonal lattice. It is a semimetal tre of the magnet will be less affected by gradient spoiling than that conducts electricity. Graphite is very commonly used in spins distant to the isocentre. pencils, under the name ‘lead’. It is found naturally in three Related Article: Gradient field forms; flake, lump and amorphous, and it can also be produced synthetically. Grand-daughter radionucleus Medical Applications: Graphite can be used as an alternative (Nuclear Medicine) In a decay chain the initial radioisotope disin- material to water in calorimeters used to measure the absorbed tegrates into another radioactive isotope. The second radioisotope dose from radiotherapy linear accelerators. is called the daughter radioisotope or decay product. The daugh- Related Articles: Carbon, Calorimeter, Dosimeter, Lattice, ter radioisotope will eventually decay to a second decay product, Radiotherapy, Water calorimeter or ‘grand-daughter’ radioisotope. This chain will continue until one of the decay products is stable. In nuclear physics, a specific GRASE (gradient and spin echo) part of a decay chain can be of clinical and/or research interest (Magnetic Resonance) See Gradient and spin echo (GRASE) (the parent radioisotope can be chosen subjectively). For instance 99Mo disintegrate to the meta-stable state of 99mTc. 99mTc decays into the grand-daughter radionuclide 99Tc. 99Tc is also radioactive Grating interferometry but emits no γ rays suitable for imaging, neither does 99Mo. It is (Diagnostic Radiology) Grating interferometry (GI) or Talbot therefore important to elute carrier-free 99mTc to avoid unneces- interferometry, is an interferometric phase-contrast imaging tech- sary radiation dose contribution to the patient. nique making use of periodic structures to prepare and analyse The relationship between parent and grand-daughter activation the x-ray beam. GI has been widely used in research and it is a is described by the Bateman equations. promising technique for future clinical implementation given its Related Articles: Parent radionucleus, Daughter radionucleus, robustness against low temporal and spatial coherence that char- Bateman equations acterise conventional x-ray sources. In a minimal GI setup, that can be used with a highly coherent Graphite (e.g. synchrotron) source, the sample is placed either between or G (General) upstream of two periodic structures (i.e. gratings), referred to as phase grating (G1) and absorption grating (G2), with a period of the order of a few micrometres (see Figure G.37). The features of G1 alter the phase of the X-ray wavefield introducing a spatially Melting point 3948 K periodic phase shift. Due to x-ray diffraction, the phase shift due Boiling point 4300 K to G1 will produce an interference pattern, or Talbot carpet, which Density near room temperature 2090–2230 kg/m3 periodically repeats at specific distances, defined as Talbot dis- Specific heat capacity 0.71 kJ/kg/K tances. In case of a phase grating of phase shift of π, the Talbot distances are: FIGURE G.37 Sketch of a grating interferometry setup. X-rays are propagating in the z-direction while the phase stepping is performed along the x-direction. The sample can be placed either between (filled shape) or upstream (dashed not filled shape) of G1 and G2. To be applied to a low-coherence source (e.g. conventional laboratory x-ray tube) the source grating G0 upstream of G1 must be added. Grating lobes 427 Grenz rays T 2 in EI each beamlet produced by the pre-sample mask is analysed ZTalbot = (n - 0.5) 4l independently from the others and no interference between differ- ent beamlets occurs. where T (>1 µm) is the half period of G1, λ (<0.1 nm) is the wave- Related Articles: Phase-contrast imaging, Edge illumination length of the x-ray beam and n is any integer. Considering realistic Further Readings: Endrizzi, M. 2018. X-ray phase-contrast parameters, the first Talbot distance (i.e. n=0) is of the order of imaging. Nucl. Instrum. Methods Phys. Res. Sect. A 878:88–98; centimetres. Pelliccia, D., M. J. Kitchen and K. S. Morgan. 2017. Theory of The absorption grating G2, which has a period T, is positioned X-ray phase-contrast imaging. In Russo, P., ed., Handbook of X-ray downstream of G1 at an integer multiple of zTalbot immediately Imaging: Physics and Technology, CRC Press, pp. 971–998; Pfeiffer, before the image receptor, and it serves to analyse the pattern F. et al. 2006. Phase retrieval and differential phase-contrast imag- produced by G1. The presence of G2 is needed, as in general the ing with low-brilliance X-ray sources. Nat. Phys. 2(4):258. detector pixels are much larger than the period T, therefore the interference pattern would not be directly resolvable. Grating lobes If one of the two gratings is laterally scanned, say along x, (Ultrasound) Grating lobes is an unwanted effect that arises as without the sample in the beam, a modulated and periodic inten- a consequence of spatial sampling of the sound beams when a sity curve, called phase-stepping curve, will be detected by each transducer array is used. Normally an array transducer should detector pixel. have only a main lobe extending perpendicular from the trans- The presence of the sample affects the interference pattern, ducer face. Grating lobes occur due to the discrete nature of the hence modifying the phase-stepping curve in three ways (see transducer elements. They can be reduced when the individual Figure G.38). The reduction of the baseline (m) describes the transducer elements are spaced less than half wavelength apart. attenuation induced by the sample, the lateral shift (φ) is linearly proportional to the refraction angle, hence to the first derivative of Gray the phase shift, and the amplitude (A) reduction describes the scat- (Radiation Protection) This is the SI unit for absorbed dose and tering (also referred to as ultrasmall-angle scattering or USAXS). kerma of ionising radiation, and is abbreviated to Gy: By making a pixel-by-pixel comparison between the phase- stepping curves acquired both with and without the sample all the 1Gy = 1J/kg three contributions can be extracted therefore yielding indepen- dently absorption, refraction and scattering maps of the imaged G The Gray was defined in 1975 in honour of Louis Harold Gray object. (1905–1965) an English physicist whose work concentrated In order to adapt GI to x-ray sources of limited spatial coher- mostly on the effects of radiation on biological systems. ence, a third grating, G0 or source grating, is added to the setup Related Articles: Kerma, Calculation of absorbed dose upstream of G1 (see Figure G.37). This (absorbing) grating divides the source into an array of individually coherent, but mutually incoherent sources. To ensure that each source produced by G0 Green’s function contributes constructively to the image-formation process G0 is (Ultrasound) A Green’s function is a solution to the inhomoge- designed such that a Talbot image is observed at G1 with the same neous differential equation: period and position. This arrangement is referred to as Talbot-Lau interferometry. Ñ2 1 ¶2g g - 2 2 = -d(r - r ) ¶ 0 As a side remark it should be mentioned that grating interfer- c t ometry is sometimes confused with the edge illumination (EI) technique, and vice versa, given the apparent similarity between where the source term on the right hand side describes a point the experimental setups. In fact, the two techniques are intrinsi- source located at r0. If the medium is unbounded and the source cally different: GI is based on the interference pattern of the peri- simple harmonic, then the field at r is g(r|r0) = eikR/(4πR), with R odically modulated x-ray radiation field produced by G1, whereas = |r − r0|, k = ω/c, ω = 2πf, and f the frequency. At a given instant, this function describes an oscillation with a period in space that is related to the ratio of the frequency of the source and the prop- agation velocity, extending outward from the source point with decreasing amplitude as 1/R. Green’s functions are used in calcu- lations of sound fields and are advantageous to use as any source function can be described as a summation of point sources. Grenz rays (Radiotherapy) The term Grenz-ray therapy is used to describe treatment with beams of very low-energy x-rays produced at potentials below 12–20 kVp. Their half-value thickness lies below 0.04 mm Al. A negligible part of the Grenz rays are absorbed in air, i.e. their half-value layer at 4 kVp is approximately 11 cm. When using a different focus-skin distance therefore the inverse square law is not so readily applicable. Another disadvantage of Grenz rays, in particular the very soft ones, is that they are so FIGURE G.38 Phase-stepping curve registered at one detector pixel strongly absorbed that it is difficult to measure them and to con- without (reference) and with (sample) the sample positioned between G trol the dose accurately. Because of the very low depth of penetra- 1 and G2. tion such radiation is no longer used in radiotherapy. Grey levels 428 Grey values Grey levels log10 L ( j ) (General) A digital image is displayed using a grid of picture ele- ments or pixels. In nuclear medicine each pixel contains the total 4 a + c × ln ( j ) + e ×(ln ( j ))2 + g ×(ln ( j ))3 + m × (ln ( j ) number of counts at a particular x–y location and is displayed at ) = 2 1+ b × ln ( j ) + d × (ln ( j )) + f ×(ln ( j ))3 + h × ( 4 5 a brightness level appropriate to this value. In black-and-white ln ( j )) + k × (ln ( j )) or greyscale displays the brightness is represented by varying degrees of white over a number of grey levels. where a to k are constants. The inverse allows discreet JNDs to be An image display system is characterised by the number of calculated over a specific range of luminance values, grey levels it can display. An 8-bit greyscale, for example can potentially display 256 (28) different grey levels. This is limited 2 however by the capability of the human eye to distinguish between j (L ) 3 = A + Blog10 (L ) + C (log10 (L )) + D (log10 (L )) the levels and the limits imposed by image noise. +E ( 4 ( 5 6 Related Article: Image display log10 (L )) + F log10 (L )) + G (log10 (L )) Greyscale 7 8 +H (log10 (L )) + I (log10 (L )) (Ultrasound) Greyscale describes the range of greys used for conventional ultrasound images where, by convention, higher intensity echoes are shown as brighter greys. The 2-D image is where A to I are constants. known as the B-mode (B for brightness) or the grey (US gray) The GSDF ensures that one-step in the JND index results in a scale image (Figure G.39). Very weak echoes are shown as black luminance difference that is a ‘just noticeable difference’, mean- and high intensity echoes as white. The distribution of grey levels ing an average human observer can perceive the luminance differ- in the image should be optimised to show good contrast resolu- ence. According to DICOM specifications, luminance can range tion of the tissue under investigation. The appearance of the grey from 0.5–4000 cd/m2. Barten’s human visual response predicts levels in the image depends on a number of instrument settings 1023 JNDs within this luminance range. j within the GSDF func- including power, gain, dynamic range, post-processing and har- tion is hence set from 1 to

1023, allowing 1023 luminance levels monic imaging. to be calculated that are approximately perceptually linearised. Related Articles: DICOM (Digital imaging and communica- G Greyscale Standard Display Function tions in medicine), Image display, Human visual response func- (Diagnostic Radiology) The Greyscale Standard Display tion, Medical image display, Barten model Function (GSDF) aims to utilise the full range of digital values Further Readings: Barten, P. G. J. 1992. Physical model input to a display device when displaying a grayscale image. If for the contrast sensitivity of the human eye. Proc. SPIE 1666, the luminance levels associated with different digital input val- Human Vision, Visual Processing, and Digital Display III. ues are perceptually indistinguishable, they are wasted, and if doi: 10.1117/12.135956; National Electrical Manufacturers luminance levels are too diffuse, an observer may see contours. Association. 2011. Digital imaging and communications in medi- Furthermore, if a system conforms to the GSDF, it is ‘percep- cine (DICOM) part 14: Grayscale standard display function. tually linearised’ when observing the ‘standard target’ (a 2-deg NEMA; Ramponi, G. and A. Badano. 2017. Method for adapting × 2-deg square filled with a horizontal or vertical grating with the grayscale standard display function to the aging eye. J. Digit. sinusoidal modulation of 4 cycles per degree). Perceptual lineari- Imaging 30: 17–25. doi: 10.1007/s10278-016-9900-2 sation ensures variations in digital input values (displayed as a specific luminance) are approximately linear as perceived by the Grey values human eye. (Diagnostic Radiology) The greyscale is the range of colour or The GSDF relates the luminance, L (in candelas per square brightness extending from black to white. In a digitised image metre, cd/m2) to the just noticeable difference (JND) Index, j, this scale is divided into discrete intervals with each interval rep- resented by a digital value. These are the pixel values corresponding to the brightness of that specific pixel. The number of different values or shades of grey a pixel can have is determined by the number of bits, i.e. the bit depth, associated with each pixel. If a digital imag- ing system uses 8 bits per pixel (depth), it can have 28 = 256 values or shades of grey or brightness with pixel values rang- ing from 0 to 255. Generally, 256 shades of grey are consid- ered adequate for viewed images. Many contemporary digital systems in medicine will capture or reconstruct images with a 12-bit depth or 4096-pixel values. This is the dynamic range of the imaging system. Windowing is then used to select a seg- ment from within this wide range for display, perhaps using around 256 values. Historically the selection of 4096 grey levels has been based on the early CT scanners, where the difference between CT numbers of air and water has been accepted as 1000, while the most absorbent bones are up to three times this absorption differ- FIGURE G.39 B-mode image of a kidney. The scale shown on the right ence, thus forming a CT number scale from -1000 (air) through 0 of the image shows the gradation of the greyscale. (water) to +3000 (bone with high absorption), that equals to 4000 Grid, Bucky 429 Grid ratio (in fact 4096 = 212). The practice has shown that 4096 levels of A major advancement was made by Dr. Hollis Potter in 1920 grey are also sufficient for various densitometric measurements who developed a method for moving the grid during the exposure (measurement of the optical density of the pixel, corresponding to to blur out the undesirable images of the grid lines. the radiation absorption of the respective voxel from the anatomi- This was known as the Potter–Bucky diaphragm. Different cal object). designs of such grids had been later introduced, as for example in CT scanners are regularly calibrated to have accurate CT num- the Lisholm scull stand, etc. bers (Hounsfield units, HU), corresponding to the respective x-ray The use of the names has changed over the years. Now the attenuation of the tissue voxels. However, other medical imaging ‘Bucky diaphragm’ is most commonly known as a grid and the systems have increased levels of noise and less accurate calcula- ‘Potter–Bucky diaphragm’ is known as a ‘Bucky’ or a Bucky tion of pixel values, compared with CT scanners. This is true for mechanism. CBCT and especially for digital radiography, where the image Related Articles: Focused grid, Grids crossed, Grid efficiency quantitative information (pixel values) does not have absolute cor- respondence to the absorption of the tissues. Due to this, the term Grid control grey values is normally used in such systems. (Diagnostic Radiology) See Grid-controlled x-ray tube Related Articles: Pixel value, Bit depth, Matrix size, Grey lev- els, CT numbers Grid efficiency (Diagnostic Radiology) Grid efficiency is the combination of two Grid, Bucky factors. One is the ability to attenuate scattered radiation, which (Diagnostic Radiology) Grid is the common name of the device is its purpose, and the other is the undesirable attenuation of the that is used in x-ray imaging to selectively attenuate the scattered primary radiation. Both of these are related to several factors but radiation between the patient’s body and the image receptor as the predominant factor affecting grid efficiency is grid ratio as illustrated on Figure G.40 (the device is also known as anti-scatter illustrated on Figure G.41. grid). Grid ratio (usually with values from 5:1 to 16:1) is a ratio The grid includes thin lead strips (absorbent lamellas) sepa- between height (thickness) of the strips and width of the inter- rated with non-absorbent plastic (or carbon fibre) material. The space. In general, the efficiency of a grid increases with grid ratio orientation of the lead strips allows for the primary x-ray beam because the attenuation of the scattered radiation increases more to pass through with minimal absorption, while the majority of than the attenuation of the primary radiation. the scattered x-rays are absorbed, as the direction of their path- Related Articles: Bucky diaphragm, Grid G ways differs significantly from the primary x-ray beam. This way mainly primary x-rays reach the detector (film). Grid, focussed Usually the lead strips have linear or crossed construction. (Diagnostic Radiology) See Focussed grid The more often used linear grids have one set of parallel strips (parallel grids and focused grids). Most of the articles and dia- Grid ratio grams here refer to the linear grids. Other grids use two set of (Diagnostic Radiology) The grid ratio of an anti-scatter grid is crossed strips (crossed grids and rhomboid grids). a design characteristic of a grid as illustrated in Figure G.42. Bucky diaphragm is the classic name for what is more com- Higher ratio grids have increased attenuation of scattered radia- monly known as a grid used to selectively absorb scattered tion and produce improved image contrast at the cost of a reduced radiation in x-ray imaging. It is named for Dr. Gustave Bucky of sensitivity. Germany who developed the first grid or ‘diaphragm’ in 1913. The grid ratio is given in values as 5:1, 12:1, etc. These values are usually given in grid specification together with the density of the lead strips (or lamellas) – for example 40, 28 L/mm, etc.). This way one grid specified as Pb 12/40 means that it has lead strips Grid function with density of 40 L/mm and grid ratio 12:1. Related Article: Grid efficiency Absorb scattered radiation Ideal grid 1.0 0.8 Primary radiation Real grid 0.6 Let primary pass 0.4 0.2 Scatter Real Grid Ideal 2 4 6 8 10 12 14 16 Grid ratio FIGURE G.40 Grid reducing scattered radiation concept. (Courtesy of FIGURE G.41 Illustration of grid efficiency. (Courtesy of Sprawls Sprawls Foundation, www .sprawls .org) Foundation, www .sprawls .org) Grid penetration Grid-control 430 Ground connection removes scatter similarly to a linear grid with ratio 12:1 (while using similar radiation dose). Related Article: Grid, Bucky Further Reading: Thompson, M., M. Hattaway, D. Hall and S. Dowd. 1994. Principles of Imaging Science and Protection, W.B. Saunders Company, Philadelphia, PA. Gross tumour volume (GTV) (Radiotherapy) The gross tumour volume (GTV) describes the Grid ratio = t/d full extent of visible malignant growth to be treated with radio- Patient therapy. It can be determined using many different imaging tech- t niques such as CT and MRI. As the different imaging techniques vary in their specificity, often many imaging techniques will be d used together with image fusion/registration techniques. Margins will be added to the GTV to form the clinical target volume Grid (CTV) and the -planning target volume (PTV). In some cases, such as post-operative radiotherapy, there may be no GTV defined Receptor (Figure G.43). The use of GTV as a planning volume was proposed by the FIGURE G.42 Grid ratio concept and formula. Some typical values ICRU in Report 50 (with addendum 62). This report provides a could be t = 2 mm, d = 0.25 mm, producing grid ratio 8:1 (the thick- common framework on prescribing, recording and reporting ness of the lead strips is on the order of 0.05 mm). (Courtesy of Sprawls therapies, with the aim to improve the consistency and inter-site Foundation, www .sprawls .org) comparability. It details the minimum set of data required to be able to adequately assess treatments without having to return to the original centre for extra information. Further Reading: Dowsett, D. J., P. A. Kenny and R. E. Abbreviation: GTV = Gross tumour volume. G Johnston. 1998. The Physics of Diagnostic Imaging, Chapmann Related Articles: ICRU, Clinical target volume (CTV), & Hall Medical, London, UK. Planning target volume (PTV), Treated volume, Irradiated volume Further Readings: ICRU. 1993. Prescribing, reporting Grid-control and recording photon beam therapy. Report 50, International (Diagnostic Radiology) See Grid-controlled x-ray tube Commission on Radiation Units and Measurements, Washington, DC; ICRU. 1999. Prescribing, recording and reporting pho- Grid-controlled x-ray tube ton beam therapy (Supplement to ICRU Report 50). Report 62, (Diagnostic Radiology) The grid-controlled x-ray tube is used International Commission on Radiation Units and Measurements, for creation of sequences of very short x-ray pulses (as for the Washington, DC. cine-radiography). This is achieved by applying small negative voltage to the Wehnelt electrode (the cathode cup). Increased negative charge of the Wehnelt electrode creates strong space- Ground connection charge effect, what blocks the thermal electrons and stops the (General) See Grounding anode current (in some tubes the Wehnelt electrode is maintained charged at about −2 kV). Applying control pulses to the Wehnelt electrode leads to very effective control of the beam of thermal electrons – hence producing sequences of 1 ms x-ray exposures. These grid-controlled tubes are rarely used at present, as the new medium-frequency high voltage generators control very effec- tively the short exposures. The name of this tube is probably related (by analogy) with the electron vacuum tubes (triodes, etc.), where a special electrode (a metal grid) is used to control the anode current. Related Articles: Cathode, Space-charge effect, Filament cur- rent, Wehnelt electrode Gross tumour volume Clinical target volume Grids, crossed (Diagnostic Radiology) In comparison to the linear anti-scatter Planning target volume grids, the crossed grids (also known as crisscross) have two set Treated volume of crossed lead strips. These can be perpendicular or at various angles one to the other (e.g. rhomboid grid). Crossed grids are Irradiated volume more expensive, but at the same time very efficient at removing scatter radiation. For example, a crossed grid with grid ratio 5:1 FIGURE G.43 Definition of target volumes as in ICRU 50. Ground lead 431 Gyroscopic radiosurgery ZAP-X Ground lead (General) The lead that provides the electrical connection to Protons in Neutrons Natural Earth (see the eponymous article). the in the Nucleus Gyromagnetic Abundance ω at 1.5 Element Nucleus Nucleus Spin I Ratio γ (s−1T−1 Related Article: Earthing ) (%) T (MHz) H1 1 0 ½ 2.675 × 108 99.985 63.864 Grounding H2 1 1 1 2.222 × 108 0.015 9.803 C12 (General) Grounding is a term used in US electrical engineer- 6 6 0 0 98.89 0 C13 ing to represent electrical equipment that has direct physical con- 6 7 ½ 6.733 × 108 1.11 16.058 nection to the ground/Earth, thus providing protection against N14 7 7 1 1.933 × 108 99.63 4.613 N15 electrical shock. In the United Kingdom the equivalent term is 7 8 ½ 2.711 × 108 0.37 6.471 O16 ‘Earthing’. 8 8 0 0 99.759 0 O17 8 9 5/2 3.595 × 108 0.037 8.658

F19 9 10 ½ 2.517 × 108 100 60.081 GSD (genetically significant dose) Na23 11 12 3/2 7.076 × 108 100 16.893 (Radiation Protection) See Genetically significant dose (GSD) P32 15 16 ½ 1.085 × 108 100 25.898 GSDF (Greyscale Standard Display Function) (Diagnostic Radiology) See Greyscale Standard Display Function Gyroscopic radiosurgery ZAP-X (Radiotherapy) The ZAP-X is a self-contained, self-shielded, 2.7 GSO (gadolinium orthosilicate) MV linac which does not typically require its own radiation bun- (Nuclear Medicine) See Gadolinium orthosilicate (GSO) ker. It is intended for the stereotactic radiosurgery (SRS) treatment of benign and malignant intracranial and cervical spine lesions. GTV (gross tumour volume) The linac is mounted within a combination of yoked gimbals with (Radiation Protection) See Gross tumour volume (GTV) attached radiation shielding, each of which accurately rotates around a common isocentre. This mechanical set-up enables the Gyromagnetic ratio radiation beam to be targetted from two pi steradians of solid angle. (Magnetic Resonance) Nuclear constant γ describing the propor- The patient is positioned on a robotic couch that extends outside G tionality between frequency and B0 in the Larmor equation: the treatment sphere but which itself is also enclosed by addi- tional radiation shielding during radiosurgery. Pairs of non-coaxial B kV x-ray images and image-to-image correlation are utilised to f 0 0 = g 2p determine the location of the patient’s anatomy with respect to the machine isocentre, both prior to and during radiosurgical treatment. Usually referring to angular frequency and thus given in radians/ Related Articles: Radiosurgery, Linear accelerator, Robotic second/Tesla. linacs In the following table a list of isotopic composition, spin, gyro- Further Reading: journalWeidlich, Georg A. et al. 2017. magnetic ratio, natural abundance and precession frequency at 1.5 Self-shielding analysis of the zap-X system. Cureus 9(12):e1917. T are reported for some nuclei in biological systems. doi:10.7759/cureus.1917 H H and D curve Peak bold effect: it typically occurs after around 5 s. If the (Diagnostic Radiology) The H and D curve, also known as the stimulus was of long duration (as in many simple block designs) a characteristic curve for radiography film, is the graph represent- plateau would be expected instead of a peak. ing the relationship between optical density and exposure. It is Negative gradient: after the cessation of the stimulus, the named for the Swiss chemist Ferdinand Hurter (1844–1898) and oxyhaemoglobin to deoxyhaemoglobin ratio begins to relax back English chemist Vero C. Driffield (1848–1915) who developed towards its baseline level. this concept for describing the characteristics of photographic Undershoot: an effect believed to stem from the fact that emulsions. increased venous blood volume normalises at a slower rate than A specific feature of the H and D or characteristic curve is increased venous blood flow, causing a relative increase in the the logarithmic scale used for exposure. The slope of the curve at deoxyhaemoglobin level. each point represents the contrast transfer characteristics of the Related Articles: fMRI (functional magnetic resonance imag- film. For more detail see the article on Characteristic curve. ing), Oxyhaemoglobin, BOLD, Block design Related Article: Characteristic curve Further Reading: Amaro, E., Jr. and G. J. Barker. 2006. Study design in fMRI: Basic principles. Brain Cogn. 60(3):220–232. Hadron therapy Epub 2006 Jan 19. Review. (Radiotherapy) Hadron therapy involves the use of proton, neu- tron or ion (e.g. carbon) treatment beams. Charged particle beams Half acquisition single-shot turbo spin echo (HASTE) have a depth dose dependence characterised by the Bragg peak, (Magnetic Resonance) HASTE is a special form of fast spin due to the dominance of collision energy losses described by the echo sequence, which combines single shot imaging with par- Bethe–Bloch equation. This means that, for a certain ion energy, tial Fourier encoding. The latter exploits the symmetry of the the dose distribution rises to a sharp maximum at a certain depth. raw data in k-space requiring the acquisition of part of the Often, the energy of the beam is modulated to smear-out the k-space data. This increases the temporal resolution of the pulse dose-depth profile, producing a spread-out Bragg peak (SOBP). sequence. Neutrons (being neutral in charge) have a depth dose character- Related Articles: Fast spin echo (FSE), Half Fourier imaging H istic similar to x-ray beams. Hadron beams typically have higher (HFI), Rapid acquisition relaxation enhancement (RARE) relative biological effectiveness (RBE) values than x-ray or elec- tron beams. Half Fourier imaging (HFI) Abbreviations: SOBP = Spread-out Bragg peak, RBE = (Magnetic Resonance) One common goal in magnetic reso- Relative biological effect. nance imaging (MRI) is to reduce the image acquisition time. Related Articles: Charged particle therapy, Hadron therapy, Conventional MRI images are generated by a two-dimensional Ion therapy, Neutron therapy, Heavy particle beams, Spread-out Fourier-transform of the k-space data. Under the assumption that Bragg peak (SOBP), Range shifter, Intensity modulated proton the image is a real function, the sampled k-space contains redun- therapy (IMPT), Passive beam scattering, Energy selection sys- dant information. tem, Pencil beam scanning (PBS), Range straggling, Relative Consider the Fourier transform connecting the image space biological effectiveness, Cyclotron, Synchrotron, Single room and the k-space, given by particle therapy systems, Bragg peak, Magnetic beam steering, Degrader, Range straggling, In vivo range verification, Organ at S(k) = òr(x)exp( - i2pk × x)dx (H.1) risk, Stoichiometric calibration, Proton arc therapy, Pion therapy, Carbon ion therapy where S(k) is the signal at the k-space position k = (kx, ky)Haemodynamic response function r(x) is the image intensity at the image position x = (x, y) (Magnetic Resonance) In functional magnetic resonance imag- ing the temporal variation of the BOLD signal associated with increased neuronal activity is known as the haemodynamic If r(x) is real, then response function. A typical haemodynamic response function for a brief stimu- S(k) = S * (-k) Û òr(x)exp( - i2pk × x)dx lus is shown in Figure H.1. The main features are as follows: ( * (H = òr(x)exp(i2pk × x)dx) .2) Dip: some studies have reported an initial dip, whilst others have failed to replicate the finding. This remains a controversial issue. since r*(x) = r(x), where r*(x) is the complex conjugate of r(x). Positive gradient: the MR signal rises due to an increase in the In other terms, the k-space contains redundant information and oxyhaemoglobin to deoxyhaemoglobin ratio. a full-resolution image could be reconstructed by only sampling 433 Half Fourier turbo spin echo 434 Half-life of radionuclides in medicine ~5 s Peak (‘BOLD effect’) 100 0.5 at 3% 75 50 ‘Dip’ ‘Undershoot’ Time (s) 25 FIGURE H.1 Haemodynamic response function from a hypothetical short duration stimulus (thick grey bar on Y axis); the BOLD effect peaks after circa 5 s from the start of the stimulus. (Image courtesy of Amaro and Barker.) 0 0 0.5 1 1.5 2 2.5 3 3.5 mm Al half of k-space and reconstructing the other half by Equation H.2. Omitting to sample parts of the k-space as described ear- FIGURE H.2 A typical curve of HVL measurement of an x-ray tube. lier is a technique called half-Fourier imaging or partial-Fourier The 50% line shows an HVL of 2.65 mm Al equivalent. imaging. However, if the image is complex (e.g. due to the influence of B0 inhomogeneities), Equation H.2 does not hold and reconstruct- HVL can be used as indicator for the total filtration of an ing artefact-free images from partial-Fourier imaging requires x-ray tube (Figure H.2). Special tables and graphs (x-ray specific) more advanced reconstruction schemes. Various methods for are used to transfer the HVL value into total filtration (mm Al doing the reconstruction have been suggested, for homodyne and equivalent). projection onto convex subsets reconstruction. For details regard- ing these methods, see references. Half wave rectification Related Article: k-space (Diagnostic Radiology) See Rectifier Further Readings: Cuppen, J. and A. van Est. 1987. Reducing MR imaging time by one-sided reconstruction. In: Topical Half-life of radionuclides Conference on Fast MRI Techniques, 15–17 May, Cleveland, OH; (Nuclear Medicine) The half-life of a radionuclide is the time Liang, Z., F. Boada, R. Constable, E. Haacke, P. Lauterbur and H taken to reduce its activity to half of its initial value. The half-life M. R. Smith. 1992. Constrained reconstruction methods in MR is unique for a given radionuclide and is usually denoted by T imaging. Rev. Magn. Reson. Med. 4: 67–185; Margosian, P. and ½. If initially there are N0 nuclei in a radionuclide, then at time t F. Schmitt. 1986. Fast MR imaging: Imaging with half the data. = T½ there is 50% of N0 nuclei remaining: Health Care Instrum. 86:195–197; Noll, D. C., D. G. Nishimura and A. Macovski. 1991. Homodyne detection in magnetic reso- nance imaging. IEEE Trans. Med. Imaging 10:154–163. N Nt = 0 2 Half Fourier turbo spin echo T ½ is related to the decay constant λ by (Magnetic Resonance) See Half acquisition single-shot turbo spin echo (HASTE) 0.693 l = T1/2 Half value layer (HVL) (Diagnostic Radiology) Half value layer (HVL) is the most fre- The half-life of a radionuclide can therefore be used to calculate quently used quantity or factor for describing both the penetrating its activity at any given time, provided the initial activity of the ability of specific radiations and the penetration through specific radionuclide is known. objects. HVL is the thickness of material penetrated by one half Further Readings: Cherry, S., J. Sorenson and M. Phelps. 2003. of the radiation and is expressed in units of distance (mm or cm) Physics in Nuclear Medicine, 3rd edn., Saunders, Philadelphia, referred to that specific material. PA; Ramesh, C. 2004. Nuclear Medicine Physics: The Basics, 6th Increasing the penetrating ability of a radiation increases its edn., Lippincott Williams & Wilkins, Philadelphia, PA. HVL. HVL is related to, but not the same as, average photon range. There is a difference between the two because of the expo- Half-life of radionuclides in medicine nential characteristic of x-ray attenuation and penetration. The (Nuclear Medicine) The half-life of a radionuclide is the time specific relationship is taken for the activity to decay to half of its original value. The ideal half-life for a radionuclide depends on the pur- HVL = 0.693´ Average range = 0.693/m pose, i.e. if the radionuclide is used for imaging or radiother- apy. Radionuclides with a short half-life are preferable when This shows that the HVL is inversely proportional to the attenu- imaging, since most of the activity will decay during image ation coefficient. The number, 0.693, is the exponent value that acquisition and no unnecessary absorbed dose is received by gives a penetration of 0.5. the patient. The principal limitations for using short-lived % MR signal change % HAMA (Human anti-mouse antibody) 435 Hard pulses radionuclides are production and quality assurance related. In antibodies can increase the likelihood of poor monoclonal anti- examinations using short-lived radionuclides the patient must body binding and allergic reactions in subsequent murine anti- receive a live feed of newly produced radionuclides from an body administrations. accelerator or cyclotron, otherwise the activity will decay while Recently, human monoclonal antibodies have been developed transporting from the production site to the examination room. using in-vitro techniques. Monoclonal antibodies produced using The live feed does not allow any complicated chemical label- these methods do not suffer from the drawbacks related to the ling or extensive quality control of the radiopharmaceuticals. HAMA response. Another factor to consider is the redistribution rate, i.e. how Related Articles: Monoclonal antibodies, Anti-idiotype anti- long until the radionuclide enters the biological process of body technique interest and for how long the process goes on. The radionuclide Further Readings: Keenan, A. M., J. C. Harbert and S. M. used for clinical imaging is therefore an optimisation between Larson. Monoclonal antibodies in nuclear medicine. J. Nucl. Med. patient dose, the biological redistribution rate, production and 26:531–537; Zhu, Z. and L. Yan. 2011. Next generation of anti- quality assurance limitations. body therapy for cancer. Chin. J. Cancer. 30(5)(May):293–302. For radionuclides used for therapy purposes there are a num- ber of extra variables to take into consideration. For example, the redistribution time for large proteins is generally in the order of Hard pulses 1–2 days and therefore these need to be labelled with a radionu- (Magnetic Resonance) To produce an image, the MRI system clide with corresponding half-life. However, as for imaging the must first stimulate hydrogen nuclei in a specific 2D image plane technical aspects prevent the use of any radionuclide

with half- in the body, and then determine the location of those nuclei within life less than ∼1 h. that plane as they precess back to their static state. These two tasks are accomplished using gradient coils which cause the magnetic field within a localised area to vary linearly HAMA (Human anti-mouse antibody) as a function of spatial location. The resonant frequencies of the (Nuclear Medicine) An antibody produced in humans following hydrogen nuclei are spatially dependent within the gradient. the administration of monoclonal antibodies manufactured from Varying the frequency of the excitation pulses controls the immunised mice. area in the body that is to be stimulated. The location of the stim- Antibody molecules, also known as immunoglobulins, are ulated nuclei as they precess back to their static state can also produced by the B lymphocyte cells of the immune system of be determined by using the emitted resonant RF-frequency and animals in response to the introduction of foreign substances or phase information. antigens. Antibodies are able to recognise and bind to specific A pulse is a rapid change in the amplitude of an RF signal or in sites on the surface of antigens, causing a complex immunological some characteristics an RF signal, e.g. phase or frequency, from response usually resulting in antigen destruction. a baseline value to a higher or lower value, followed by a rapid H An antigen may have multiple binding sites and B lympho- return to the baseline value. cytes will therefore differentiate to produce a range of antibod- For radio frequencies near the Larmor frequency, it will ies which are able to bind specifically to each site. Monoclonal result in rotation of the macroscopic magnetisation vector. The antibodies are produced by removing sensitised B lymphocyte amount of rotation will depend on the strength and duration of cells (i.e. cells which have been exposed to the desired anti- the RF pulse; commonly used examples are 90° (π/2) and 180° (π) gen) from an immunised animal and extracting and cloning pulses. one specific B lymphocyte. This is achieved using a process RF pulses are used in the spin preparation phase of a pulse called hybridoma technology. Each cell clone has the potential sequence, which prepare the spin system for the ensuing to produce a single species of antibody, otherwise known as a measurements. monoclonal antibody, which will exclusively bind to a site on In many sequences, RF pulses are also applied to the volumes the desired antigen. outside the one to be measured. This is the case when spatial pre- Radiolabelled monoclonal antibodies are a useful tool to detect saturation techniques are used to suppress artefacts. the presence of a specific antigen in diagnostics, or to treat cancer- Many preparation pulses are required in MR spectroscopy to cell specific antigens. 99mTc labelled sulesomab is an example suppress signal from unwanted spins. of a monoclonal antibody that can be utilised to image infection The simplest preparation pulse making use of spectroscopic and inflammation. In therapeutics, 90Y-ibritumomab tiuxetan properties is a fat saturation pulse, which specifically excites the (Zevalin) and 131I-tositumamab (Bexxar) are two radioisotope fat resonant frequency, so that the magnetisation coming from conjugated anti-CD20 monoclonal antibodies developed for the fat protons is tilted into the xy-plane where it is subsequently treatment of non-Hodgkin’s lymphoma. destroyed by a strong dephasing gradient. Historically, mice have been used for immunisation and cell The frequency spectrum of RF pulses is critical as it deter- extraction. Monoclonal antibodies produced from murine B lym- mines the spatial extension and homogeneity over which the spin phocytes are foreign to humans and therefore commonly cause magnetisation is influenced while a gradient field is applied. the production of Human Anti-Mouse Antibodies (HAMA) There are many RF pulse shapes used in MRI. The most com- when introduced into humans. HAMA can interact directly with monly used are sinc pulse, Gaussian pulse and rectangular pulse. the desired antigen binding site, therefore acting as an anti-idio- Gaussian and sinc pulses are mostly used in the presence of an type antibody, reducing the binding efficacy of the monoclonal applied field gradient for slice-selective excitation. antibodies. When the target is to excite an entire volume, a very brief, HAMA can also cause allergic reactions in humans and a strong, rectangular pulse (often called a ‘hard’ pulse) is used with HAMA response following initial administration of murine no gradients applied. Hardening agent 436 Head, pitch and roll imaging these multiple reflections turn up as a ‘haze’ in the image. However, the acoustic pressure does not build up here to the levels where harmonics are generated, and this unwanted noise is sup- pressed in harmonic mode. In other words, the contrast resolution has improved. HASTE (half acquisition single-shot turbo spin echo) (Magnetic Resonance) See Half acquisition single-shot turbo spin echo (HASTE) Hazard (Radiation Protection) Hazard, or health hazard, is defined as anything that has the capacity to cause harm to a human being. Such harm can be through physical injury, biological damage or other health detriment (adverse effect) in the form of cancer or other disease. Examples of hazards include fire, chemicals, electricity (electrocution), water (drowning), and ionising and non-ionising radiation exposure. If the hazard relates to radiation exposure, it may be called a radiation hazard. Related Articles: Adverse effects, Risk assessment Hazard distance (HD) and Hazard value (HV) (Non-Ionising Radiation) The hazard distance of a broadband artificial optical radiation source is the distance at which the level Related Article: RF pulses of exposure is equal to the relevant exposure limit value (ELV). Further Readings: Collins, Christopher M. Fundamentals Beyond this distance the ELV is not exceeded and there is no risk of MRI—fields and basic pulse sequences; Scampin, John. of injury. This distance is often used to set boundaries within Introduction to Magnetic Resonance Imaging (MRI); Wang, J., which control measures are required for the use of the source. W. Mao, M. Qiu, M. B. Smith and R. T. Constable. Factors influ- The hazard value (HV) is a different way of presenting this encing flip angle mapping in MRI: RF pulse shape, slice-select information. The HV is given by Equation 1 and is the ratio of the H gradients, off-resonance excitation, and B0 inhomogeneities. exposure level at a specific distance to the ELV at that distance. So, if HV is greater than one the ELV is exceeded and control Hardening agent measures are required. If the HV is less than one, the ELV is not (Diagnostic Radiology) A hardening agent is a chemical compo- exceeded at that distance and for a specific exposure time. nent of the fixer solution used in radiographic film processing to shrink and harden the emulsion layer after the processing actions HV (distance, exposure time) are complete. Aluminium chloride is typically used as a hardener. Exposure Level (distance, exposure time) (H.3) = Exposure Limit Value (ELV) Hardening, beam (Diagnostic Radiology) See Beam hardening Harmonic imaging Related Article: Exposure limit value (ELV) (Ultrasound) Harmonic imaging is an imaging modality that cap- Further Reading: A Non-Binding Guide to the Artificial italises on non-linear effects to produce images with improved Optical Radiation Directive 2006/25/EC, Radiation Protection contrast and/or resolution. The first clinical application of har- Division, Health Protection Agency. monic imaging was to improve the detection of contrast agents. The approach was simply to transmit a pulse at one (centre) HC (homogeneity coefficient) frequency, but have the detection at twice that frequency using (Radiotherapy) See Homogeneity coefficient (HC) a bandpass filter. This assumes a transducer that is wideband enough, and also that the pulses need to be fairly narrowband HDR in order to avoid spectral leakage from the fundamental to the (Radiotherapy) See High dose rate harmonic. A bi-effect was also that the system was sensitive enough to Head, pitch and roll pick up the harmonics generated by non-linear distortion. The (Radiotherapy) There are three main axes about which movement idea had been suggested earlier, but had not gained clinical prac- can happen – vertical, longitudinal and lateral. The movement can tice. Harmonics are generated at high acoustic pressures (see non- be either a displacement or a rotation about the axis. When the linear propagation), which naturally is in the centre of the sound motion is a rotation, the terms head, pitch and roll are used for beam. Thus the beam of the detected sound is narrower than the the three axes as illustrated in Figure H.3. Occasionally the term transmitted, and consequently the lateral resolution is improved. ‘head’ is replaced by ‘yaw’. These terms can be used in the radio- An advantage in cardiac investigations is also that multiple reflec- therapy context when checking the gantry set-up of a linac or for tions between the transducer face and the ribs, for instance, are the rotational set-up error of a patient. not detected in harmonic imaging mode. In normal (fundamental) Related Article: Set-up error Head coil 437 H eating Vertical or other disease. Examples of hazards include fire, chemicals, ‘head’ electricity (electrocution), water (drowning) and ionising and non-ionising radiation exposure. If the hazard relates to radiation exposure, it may be called a radiation hazard. Related Articles: Adverse effect, Risk assessment Health Protection Agency (HPA) Lateral (Radiation Protection) In 2005 the National Radiological ‘pitch’ Protection Board merged with the Health Protection Agency (HPA) to become an independent body that assesses the impact of all sources of radiation on public health in the United Kingdom, and advises on the safe use of radiation. The HPA also monitors communicable diseases, performs Longitudinal epidemiological studies and advises on infection and chemical ‘roll’ hazards. The HPA has three core centres: Centre for Emergency Preparedness and Response, Centre for Radiation, Chemical and FIGURE H.3 An illustration of the three axes and the relevant term Environmental Hazards and Centre for Infections. relating to rotation about that axis. The HPA runs various training programs for health profes- sionals and carries out simulations to test public health systems in times of emergency. Heat unit (Diagnostic Radiology) Heat unit (HU) is a practical measure introduced in the past to measure the heat capacity of the x-ray tube anode. This measure continues to be used, but as it is directly related to energy, it is often replaced by Joules (J). By definition 1 HU = 1.4 J. HU has been introduced when most high-voltage generators have been single phased two pulsed type. Due to this reason HU refers to the effective kV (i.e. the root mean square kV = 0.71 kVp) and the average mAs. Effective kV (or r.m.s. kV) is used to represent the constant kV with the same effect as the original H pulsating kVp. This way the HU for such generators are calculated very easily: 1 HU = 1.4 kVeff mA s = 1.4 (0.71 kVp) mA s = kVp mAs The introduction of other types of generators (as three phased, constant potential, etc.) changed the representation of the kVp into constant kV (as the wave form pulsations are different), what requires introduction of specific coefficients related to the pulses of the kVp (i.e. the type of rectification). This way: a. HU for three phase six pulse generator (when kVp are very close to constant kV) will be FIGURE H.4 Head coil. 1HU = 1.4 kVeff mA s = 1.4 (0.96 kVp) mA s = 1.35 kVp mAs Head coil b. HU for constant potential generator (as most medium (Magnetic Resonance) A head coil is an RF coil used in brain and frequency generators) will be head imaging. Head coils are volume coil designs, providing good signal 1HU = 1.4 kVp mAs uniformity throughout the imaged volume. Head coils can be designed in a transmit/receive (T/R) configuration. Use of a T/R The heat capacity of the anode of an x-ray tube is very often pre- coil design avoids the need for RF excitation by the body coil and sented in HU, thus a good x-ray tube for computed tomography reduced SAR (specific absorption rate). will have several millions heat unit capacity (e.g. 5 MHU). The most commonly used head coils (Figure H.4) are circu- Related Article: Voltage waveform larly polarised head coils such as a birdcage coil or multiple array Further Reading: Dendy, P. and B. Heaton. 2002. Physics for head coils such as an eight channel coil. Diagnostic Imaging, IOP Publishing, Philadelphia, PA. Hyperlink: www .sprawls

.org Health hazard (Radiation Protection) Health hazard, or hazard, is defined as Heating anything that has the capacity to cause harm to a human being. (Ultrasound) As ultrasound passes through tissue, some of its Such harm can be through physical injury, biological damage energy is absorbed in tissue and is converted to heat. Ultrasound or other health detriment (adverse effect) in the form of cancer induced heating is dependent on Heating, radiofrequency 438 Heavy charged particle stopping power • Intensity Heating, radiofrequency • Tissue properties, e.g. absorption (Magnetic Resonance) See Radiofrequency heating • Heat clearance, conduction and perfusion • Proximity to the transducer Heating, resistive (Diagnostic Radiology) See Filament Heating is a potential adverse effect for diagnostic ultrasound. It is known from tissue and animal studies that elevated tempera- tures can cause biological effects and that these are dependent Heavy charged particle stopping power on temperature rise and length of time at raised temperatures. It (Radiotherapy) The stopping power may be subdivided into col- is known from in vitro studies that ultrasound at high diagnostic lision and radiative stopping power. The latter depends on the levels can raise temperatures in the order of 1°C–2°C. Studies inverse square of the mass of the particle for a given particle of outputs, the heating effect and cell and animal experiments velocity and therefore it becomes practically insignificant for par- have been used to define safe levels at which to operate. This ticles other than electrons and positrons. The stopping power for a is particularly important when scanning embryos and foetuses heavy particle in any medium is given therefore by since a largely normal population is screened at a time when rapid development occurs. Non-ultrasound heating studies have demonstrated teratogenic effects at temperature rises of 4°C for dE 4pk2z2 4 é ù 0 e n ê 2mc2b2 5 min. Effects at lower temperatures are uncertain; other factors - = ln - b2 ú dx mc2 b2 ê ë I (1- b2 ) ú may be important, for example if the mother has raised tempera- û ture herself. The World Federation of Ultrasound in Medicine where and Biology (WFUMB) have stated that systems producing a k0 = 8.99 109 N m2/C2 temperature rise of no more than 1.5°C can be used clinically z is the atomic number of the heavy particle without reservation. e is the magnitude of the electron charge Manufacturers, regulatory bodies and user groups have devel- n is the number of electrons per unit volume in the medium oped guidelines and recommendations to ensure that heating m is the electron rest mass caused by diagnostic ultrasound is low and that outputs sufficient c is the speed of light in vacuum for effective imaging are permitted. The current output safety β = v/c is the speed of the particle relative to c indices (ODS) include a Thermal Index displayed on systems as I is the mean excitation energy of the medium an indication of the relative heating risk. Ultrasound probes are themselves prone to heating. In Examination of the equation shows that stopping power is a advanced matrix arrays, the high density of elements can cause H function of the charge ze and velocity V of the heavy charged par- temperature rises that required cooling, either by conduction to ticle but not explicitly of its mass. Therefore if the stopping power the rear of the transducer case or cooling by circulating fluid. is known for one type of heavy particle of mass mP it can be found For therapeutic ultrasound devices, for example those used in for any other singly charged particle of mass mB by changing the physiotherapy, low temperature heating may be the desired mech- energy scale so that the new energy values are mB/mP times the anism used. In high intensity focussed ultrasound (HIFU) heating old (Figure H.5). For the medium only the electron density n is is used as the mechanism to destroy pathogenic tissue. The power important. and intensity are much higher than for diagnostic applications and The expression of stopping power for any heavy charged par- the design of HIFU systems is markedly different from diagnostic ticle in any medium can be written as systems and other therapeutic applications. Related Articles: Thermal index, Intensity, HIFU Further Reading: Ter Haar, G. and F. A. Duck, eds. 2000. dE 5.08´10-31z2n The Safe Use of Ultrasound in Medical Diagnosis, BMUS/British - = éF I dx b2 ë (b) - ln eV ùû MeV/cm Institute of Radiology, London, UK. 0.050 Protons 0.040 π-mesons 0.030 α particles μ-mesons 0.020 Electrons 0.010 Deuterons 0 10–2 10–1 1 10 102 103 104 Energy, MeV FIGURE H.5 The stopping power of air for different particles versus particle energy. dT/dx, MEV/CM air Heavy ions 439 Helical artefact where Heavy particle beams (Radiotherapy) Heavy particle beams are beams of neutrons, . F (b) 1 02 ´106b2 protons, light ions (such as carbon) used in hadron therapy. = ln 2 - b2 1- b Charged heavy particle beams are generated using particle accelerators. These accelerators are substantially larger than the more familiar electron linac due to the higher energies needed Related Articles: Stopping power, Collision mass stopping to penetrate tissues to the required depths and the high magnetic power, Electron stopping power, Mass collision stopping power, rigidity of the much heavy hadron particles. Neutron beams are Mass radiative stopping power, Mass stopping power, Restricted generated using three methods: Hadron accelerators with a light mass collision stopping power, Restricted stopping power target to generate neutrons; radioactive sources; and nuclear reactors. Heavy ions Related Articles: Charged particle therapy, Hadron therapy, (General) A heavy ion is one with a mass greater than an alpha Ion therapy, Proton therapy particle. They can be used in charged particle therapy. Please see related articles for more information. Heel effect Related Articles: Charged particle therapy, Hadron therapy, (Diagnostic Radiology) See Anode Heel effect Ion therapy, Proton therapy, Heavy particle beams Helical artefact Heavy metal filter (Diagnostic Radiology) Helical (spiral) CT suffers in general (Radiotherapy) Heavy metal filters are commonly used in radio- from the same artefacts as sequential (axial) CT. However, there therapy equipment in several ways, including the following. are effects, or artefacts, which are seen only using helical CT, and these are due to the spiral interpolation process which gives rise 1. Beam flattening filter – The intensity of an x-ray beam to data inconsistencies at different angular orientations. These generated in mega-voltage radiotherapy equipment artefacts will be particularly marked for objects varying rapidly such as linear accelerator has a Gaussian distribution in the z-direction, and at higher helical pitches. Figure H.6 shows across the radiation field. A conical shape metal flat- images of a cone from a single slice spiral CT scanner (Circa, tening filter of several cm in thickness of copper or 1999), starting with the thinnest part of the cone in the top left lead alloy is used to flatten the beam so as to achieve a hand corner. The helical artefact is apparent by the non-circular uniform beam intensity across the field. The flattening appearance of the images. The shading caused by the interpola- filter normally sits on a turret, which rotates to fit dif- tion process varies in angular position with the z-axis (scan axis) ferent flattening filters in position when beam energy is position of the reconstructions. changed, and is located between the x-ray target and the In multislice helical scanning this helical artefact has the H ionisation chamber. Modern megavoltage radiotherapy appearance of the vanes of a windmill (Figure H.7) and is fre- equipment such as Tomotherapy and the latest genera- quently referred to as the windmill artefact. tion of linear accelerators do not require the use of flat- The helical artefact can be reduced by use of thinner acquired tening filters. This is because uniform beam intensity slices, lower pitch values and finer z-sampling. It should be is no longer required in advanced treatment techniques noted that these images represent artefacts as seen on older spi- such as intensity modulated radiotherapy and intensity ral scanners. Current multislice spiral scanners employ refined modulated arc radiotherapy. reconstruction algorithms which result in decreased helical 2. Wedge filter – See Wedge. artefacts. 3. Hardening filter – Hardening filters are commonly Related Articles: Artefact, Beam hardening, Cone beam arte- used in x-ray therapy units operating in the kV range, fact, Helical pitch, Image artefact, Metal artefact, Motion artefact, such as superficial x-ray therapy (see Superficial ther- Partial volume effect (artefact), Ring artefact, Spiral interpolation apy). The main purpose is to increase the penetration power of the x-ray beam to meet the required treat- ment depth. The filter cuts down the amount of low energy x-ray component and this effectively increases the mean energy of the beam, but at the expenses of lower beam intensity. 4. Electron beam flattening filter – This is a thin rather than heavy metal foil made of gold or copper used for flattening radiotherapy electron beams. This type of filter is also known as electron beam scatterer. Electron flattening filter also sits on the same turret as the x-ray flattening filter for automatic selection to meet treatment requirement. The main disadvantage of using electron scatter is that it generates brems- strahlung radiation which is undesirable in electron beam treatment. The amount of bremsstrahlung radi- ation produced is proportional to electron energy. FIGURE H.6 Single slice helical CT scan of a cone: images at different In order to reduce x-ray contamination, a double z-axis positions scanned with pitch = 2. Note this image is from a scan- scatter design is used in high energy electron beam ner ∼1999. With current reconstruction techniques helical artefacts are treatment. reduced. (Courtesy Philips Healthcare, Best, the Netherlands.) Helical pitch (CT) 440 H elical scanning Further Reading: Wilting, J. E. and J. Timmer. 1999. Artefacts For faster or slower couch speeds, i.e. non-contiguous scanning, in spiral-CT images and their relation to pitch and subject mor- the pitch value increases and decreases respectively (Figure H.8). phology. Euro. J. Radiol. 9(2): 316–322. The advantage of using a high pitch is a reduction in exam time, although on single slice scanners it also results in a broader slice Helical pitch (CT) thickness or slice sensitivity profile. Another advantage of high (Diagnostic Radiology) In helical scanning the speed that the pitch on single slice scanners is dose reduction. patient support table moves through the gantry during the scan can Higher pitches result in an increase in reconstruction artefacts, be varied. The ratio of the table feed per x-ray tube rotation to the and so pitch values greater than 1.5 are seldom used. Because x-ray beam length in the z-axis is referred to as the helical pitch, of the high gantry rotation speeds and increased detector array pitch ratio, or simply, pitch. If the table feed per gantry rotation lengths on MSCT scanners, use of high pitches is rarely necessary is equal to the length of the x-ray beam, the pitch is equal to one as scan times are sufficiently short at pitch values of around one (Equation H.3). This is often referred to as contiguous scanning: or less. Related Articles: Helical scanning, Slice thickness, Image (Table feed per rotation) artefact Helical pitch = (H.4) (z-axisx-ray beam length) Helical scanning (Diagnostic Radiology) In computed tomography (CT), the terms helical scanning and spiral scanning are synonymous. Helical scanning was introduced into clinical practice in 1990. It was enabled by the implementation of slip-ring technology on CT scanners, which removed the need for power and signal cables between the stationary and rotating parts of the gantry, and so allowed continuous rotation of the x-ray tube and detectors in one direction. When this is coupled with simultaneous couch move- ment through the gantry during the scan, a volume of attenuation data is acquired. The introduction of helical scanning revolution- ised CT, as it reduced the duration of a scan, thereby decreas- ing artefacts associated with voluntary and involuntary patient motion. The filtered back projection process, commonly used to recon- struct images in CT, requires attenuation data for multiple angular H positions in the scan plane (Figure H.9). When scanning in heli- cal mode the angular data set is non-planar as the x-ray beam describes a helical path around the patient (Figure H.10). It is FIGURE H.7 Multislice helical

artefact (windmill artefact) of a Teflon rod at 60° to the horizontal. (Courtesy of ImPACT, UK, www . impactscan .org) Helical path of beam Pitch 1 Pitch 2 Pitch 0.5 z-axis Direction of couch movement FIGURE H.8 Pitch in CT scanning (from left to right: pitch 1, pitch 2, pitch 0.5). FIGURE H.10 Helical path of x-ray beam around patient. X-ray tube Gantry Patient Table Arc of z-axis detectors Side view of detector bank (a) Power data (b) FIGURE H.9 CT scanner acquiring data in helical mode (a) scan plane, (b) sagittal plane. (Courtesy of ImPACT, UK, www .impactscan .org) Helium 441 H erring bone artefact therefore necessary to interpolate the acquired data in order to neutrons. An incoming neutron reacts with the gas to form tritium obtain a planar data set for image reconstruction. and protons, which can be detected and converted to an electrical Different types of interpolation algorithms are used and signal. these give rise to different slice thicknesses (slice sensitivity or Related Articles: Magnetic resonance imaging, MRI, z-sensitivity profiles). The slice sensitivity profile is widened to a Cryogen, Geiger–Müller (GM) counters greater or lesser extent depending on the type of algorithm used. The slice sensitivity profile is also affected by the helical pitch, particularly on single slice scanners, where higher pitches lead to Helium-free magnet wider profiles. (Magnetic Resonance) So-called helium-free magnets are super- Besides its speed, helical scanning has the advantage that conducting magnets requiring a very limited amount of helium, slices can be reconstructed at any arbitrary position along the scan about 7–20 liters compared to about 1500 liters of the MRI sys- axis (z-axis). It is common to reconstruct overlapping images, i.e. tems based on conventional superconducting magnets. While con- where the reconstruction interval is less than the slice width, and ventional superconducting magnets are completely immersed in this results in improved detectability of low contrast structures, as helium; helium-free magnet coils lie in the vacuum and are cooled well as improved multi-planar and 3-D reconstructions. by thermally connected tubes. The liquid helium circulates in the A third advantage of helical scanning, available on single slice cooling tubes. The total weight of the MR unit is reduced of about scanners, is the opportunity for dose reduction. Patient dose is 1000 kg compared to the conventional superconducting scanners. inversely proportional to the pitch, although is obtained at the cost In helium-free magnets the helium is fully sealed in the system of a wider slice sensitivity profile. during manufacturing and no more refilling with liquid helium is Related Articles: Helical pitch, Slice thickness necessary during their lifetime. In case of a magnetic quench, all the helium is fully contained into the system, therefore no pipes are necessary for safe venting. Helium Related Articles: Electro-magnet, Magnet, Permanent mag- (General) net, Quenching, Resistive magnet, Superconductivity Symbol He Helium, liquid Element category Inert gas (General) See Helium Mass number A 4 Atomic number Z 2 Helmholtz coil Atomic weight 4.0026 g/mol (Magnetic Resonance) A Helmholtz coil configuration is a pair Electronic configuration 1s2 of coils used to create uniform magnetic field in the centre of Melting point 0.95 K at 2.5 MPa the space between them. The diagram shows two circular coils H Boiling point 4.22 K with radius a, and separated by a distance d with each carrying Density near room temperature 0.1786 g/L equal currents I. The resulting magnetic field is a summation of the magnetic fields generated by each coil. It can be shown that in the case where the distance d equals the radii a, the magnetic History: Helium was discovered in 1868 by Pierre Janssen and field is maximally uniform along the axis joining the coil centres. Norman Locker, who independently observed its spectral line in This configuration is called a Helmholtz coil (Figure H.11). The light from a solar eclipse. Helmholtz coil is useful in Magnetic Resonance imaging due to Isotopes of Helium: Helium occurs naturally as two stable iso- its ability to create a uniform magnetic field. topes. By far more abundant is 4He, which makes up 99.99986% of all natural helium. The remaining 0.00014% consists of stable Herring bone artefact 3He, which is a product of tritium beta decay. Six other isotopes (Magnetic Resonance) The herring bone or ‘crisscross’ artefact is exist, all of which are unstable and have half-lives of less than 1 s. caused by spike noise in the raw data, typically due to RF inter- Medical Applications: Hyperpolarised Gas Ventilation ference. When Fourier transformed, the spike produces a regular Imaging – 3He is used to view lung ventilation. The 3He gas is series of high and low intensity stripes in one or more directions polarised using laser light before inhalation, and its distribution across the image. This leads to the herring bone or crisscross in the lungs can then be imaged using low flip angle or steady effect. state magnetic resonance imaging. 3He is expensive due to its low natural abundance. Guiding Lasers: A combination of helium and neon gases I is a commonly used gain medium for guiding lasers, which produce light in the visible region of the electromagnetic spec- trum. These are used for visual alignment of other lasers (e.g. I excimer, dye, etc.) in medical procedures. Low-power helium- neon lasers are also used therapeutically in bio-stimulation and d phototherapy. Superconducting Magnet Coolant: Liquid 4He is used to a cool the superconducting magnets, such as those used in MRI scanners, to temperatures below that at which the magnet metal a becomes superconducting. Neutron Detection: 3He gas is used in gas-filled ionisation chambers such as the Geiger–Müller tube, to enable detection of FIGURE H.11 Helmholtz coil (for d = a). Hertz (Hz) 442 High contrast Hertz (Hz) must provide a life-cycle management plan that details safe use, (General) Hertz (Hz) is the base unit of frequency in the security and disposal of the source. International System of Units (Système international d’unités – SI). 1 Hz is 1 cycle per 1 s. The unit can be applied to various High contrast periodic events and is named after the German physicist Heinrich (Diagnostic Radiology) There are several terms specifying the Hertz. subject contrast (mainly used in x-ray radiography). The most Related Article: Frequency often used ones are high contrast and low contrast. High contrast (Figure H.12) is used to describe images with Heterogeneity large differences between the optical densities of the adjacent (Radiotherapy) A heterogeneity is a region in an irradiated objects. A number of synonyms are used in practice – as hard medium which has a different composition (atomic number or contrast or short scale contrast. The latter refers to the fact that density) to that of the surrounding medium. The human body is x-ray films with small latitude present a limited (short) scale of such a heterogeneous medium, being composed of a variety of grey levels. tissues and cavities with different radiological properties than that Low contrast (Figure H.13) is used to describe images with of water. small differences between the optical densities of the adjacent The presence of these heterogeneities, with attenuation and objects. A number of synonyms are used in practice – as soft scattering properties different to that of water, can affect the dose contrast or long scale contrast. The latter refers to the fact that distribution by altering both the primary and scatter contributions x-ray films with big latitude present a large (long) scale of grey to a point relative to that in a homogeneous water medium. levels. A non-water equivalent component in the human body can be An objective way to describe the contrast (optical densities) referred to as a heterogeneity, an inhomogeneity, a tissue hetero- in radiography is based on use of densitometer. The optical den- geneity or a tissue inhomogeneity. sity (OD) values obtained by this device are logarithmic values of Example: Bone and lung are two examples of heterogeneities light transmission (OD = log10[I/I0]) through the film. Usually dis- in the human body. tinctive difference between the densities of two adjacent objects Related Articles: Tissue heterogeneity, Inhomogeneity cor- is considered to be 0.2 OD (this has been determined by many rection factor subjective assessments). Most often, these 0.2 OD are achieved Further Reading: AAPM. 2004. Tissue inhomogeneity cor- by difference of 58% between the x-ray exposures received by rections for megavoltage photon beams. Report number 85, the two adjacent areas. Hence the maximal acceptable film fog Medical Physics Publishing, Madison, WI. is also 0.2 OD. Optical densities below 0.15 OD are difficult to distinguish by a normal human vision. In digital imaging, the window technique determines the type H High activity sealed source of contrast. A large window width (e.g. WW > 200) will present (Radiation Protection) High activity sealed source (HASS) many grey levels and the image will be considered to be with describes any sealed radioactive source that is covered by regula- low contrast (many anatomical structures will be seen, but with tion introduced under European Directive 2003/122/EURATOM less differences between these) – Figure H.14. Correspondingly, to provide additional safety and security in the management of small window width (e.g. WW < 100) will present few grey levels such sources. From 1 January 2006 European member states were and the image will be considered to be with high contrast (only required to register any single source exceeding the following limits as HASS: Radioisotope HASS threshold: 55Fe 400 GBq 60Co 4 GBq 75Se 30 GBq 85Kr 100 GBq 90Sr 3 GBq 109Cd 300 GBq 137Cs 20 GBq 192Ir 10 GBq 226Ra 2 GBq 241Am 100 GBq For example, in the United Kingdom, establishments are required to be authorised by the Environment Agency to hold radioactive substances. The requirements for holding medium/low activ- ity sealed sources are covered under one license that allows the organisation some flexibility to manage various sealed sources, with a broad set of activity limits and security requirements. However, a HASS must be individually registered with the agency, and requires an individual HASS license which imposes FIGURE H.12 High contrast image – radiography of the spine. additional security and management arrangements. The owner (Courtesy of EMERALD project, www .emerald2 .eu) High counting rates 443 High dose rate (HDR) FIGURE H.15 High contrast image of test object with digital fluoros- copy and small window width – 64. (Courtesy of EMERALD project, www .emerald2 .eu) FIGURE H.13 Low contrast image of the same region. (Courtesy of EMERALD project, www .emerald2 .eu.) H FIGURE H.16 High contrast resolution measurements of a CT system. (Courtesy of EMERALD project, www .emerald2 .eu) FIGURE H.14 Low contrast image of test object with digital fluoros- High counting rates copy and large window width – 512. (Courtesy of EMERALD project, (Nuclear Medicine) See Dead time www .emerald2 .eu) High dose rate (HDR) (Radiotherapy, Brachytherapy) limited anatomical structures will be seen, but their contrast will Dose Rates in Brachytherapy: Different dose rates are used be distinctive) – Figure H.15. in brachytherapy, connected to different treatment techniques. Terms used often in computed tomography (CT) practice are ICRU, the International Commission on Radiation Units and high contrast resolution (Figure H.16) and low contrast resolution Measurements, defined these dose rates in its Report No. 38 ‘Dose (or detectability) (Figure H.17). The first one refers to the measure and Volume Specification for Reporting Intracavitary Therapy in of resolution by test objects with high subject contrast between Gynaecology’: their inserts. This measure represents the spatial resolution of the system (Lp/mm). The second one refers to measures using test 1. Low dose rate, LDR objects with small subject contrast between their inserts. This a. 0.4–2.0 G/h measure is used to assess the noise and the contrast resolution of b. Traditional radium technique; 0.5 Gy/h, 60 Gy with the system (limiting contrast, %). treatment time 5 days Related Articles: Subject contrast, Low contrast, Film lati- c. Large amount of clinical data tude, Contrast resolution, Window d. (NOTE! Ultra low dose rate 0.01–0.3 Gy/h!) High energy electrons 444 High-frequency generator beam uniformity moving away from the point of final collima- tion and the change in depth dose penetration when they traverse inhomogeneities. High-frequency generator (Diagnostic Radiology) This contemporary x-ray generator gradually replaces the classical high-voltage generator. The high frequency generator uses electrical current with 5–25 kHz fre- quency and due to this reason it is better named by some

sources as medium frequency generator. The principle of this new high-voltage generator is based on the known transformer equation: U ~ An f where A is the cross section of the transformer core (mm2) FIGURE H.17 Low contrast resolution measurements of a CT system. (Courtesy of EMERALD project, www .emerald2 .eu) n is the transformer ratio (based on the number of secondary/ primary windings) f is the frequency of the secondary voltage U 2. Medium dose rate, MDR a. 2–12 Gy/h For one x-ray high-voltage transformer n is constant (usually b. More seldom used around 500). In this case it is obvious that increasing the frequency 3. High dose rate, HDR of the electricity f will allow for reducing the size of the transformer a. >12 Gy/h = 0.2 Gy/min core A classical iron-core transformers cannot work with high fre- b. Treatment times approx. 5–20 min (external beam quencies (there will be too many losses due to their low magnetic per- therapy belongs here) meability). However new ferrite-core transformers have much higher c. Clinical data available magnetic permeability and work with high frequencies. This allows 4. Pulsed dose rate, PDR great reduction of the transformer size. As a comparison, an iron- a. Mimics LDR, using many small ‘HDR pulses’ dur- core high-voltage transformer (from classical x-ray generator) will H ing a longer treatment time weigh around 300–600 kg (depending on its power), while a con- b. Example: one pulse per hour during 24 h, 0.5 Gy temporary ferrite-core high-voltage transformer will be much lighter per pulse given in 5 min; total dose 12 Gy per day. – approximately 25% of the volume and weight of the classical one. The principal diagram of a medium (or high) frequency x-ray The radiobiological effects in the tissues irradiated depend generator is given on Figure H.18. The mains power supply is on the type of applicator used, on the fractionation scheme and rectified and then an DC/AC converter (also called inverter or on both dose and dose rate distributions. As stated in the ICRU chopper) is used convert the DC voltage to a series of pulses (usu- Report 38: ‘the clinical experience accumulated with radium ally with frequency of the order of 5–25 kHz, depending on the techniques cannot be applied to new irradiation conditions with- manufacturer). This high frequency electricity is transformed to out careful consideration’. This includes consideration of both high voltage by the special ferrite-core transformer, then is again tumour effects and effects on normal tissues. rectified before supplying the x-ray tube with high voltage (also Abbreviation: ICRU = International Commission on Radiation smoothed by special capacitor). Units and Measurements. The DC/AC converter is normally thyristor-type, driven by Related Articles: Brachytherapy, Dose rates in brachytherapy, external pulse generator. This combination of medium frequency see also articles under radiobiology and converter is further applied not only to the high-voltage trans- Further Reading: ICRU. 1985. Dose and volume specification former, but also to the filament circuit, anode rotation, etc. From for reporting intracavitary therapy in gynecology. ICRU Report this point of view a contemporary x-ray generator will have a 38, Washington, DC. number of ferrite transformers. High energy electrons (Radiotherapy) High energy electron beams in the energy range DC–AC HF–HV ferrite of 4–20 MeV are useful to treat superficial tumours and some- Rectifier converter transformer Rectifier x-ray tube times they are used in conjunction with photon beams either as a boost or a mixed-beam combination. The interaction of elec- trons with matter (collision) has specific features that allow the effective irradiation of relatively superficial tumours sparing the AC ~ normal tissues beyond. In a treatment plan with photons the beam is exponentially absorbed by the tissues beyond the target depth while the electron beam, as constituted by charged particles, 5–25 kHz interacts with atoms and shows a finite range in matter limit- ing the dose to deep sited normal tissues. Disadvantages in the FIGURE H.18 Block diagram of medium (high) frequency high-voltage use of high energy electron beams are the rapid decrease of the x-ray generator. High-intensity focused ultrasound (HIFU) 445 H igh-voltage cable Although these XG are more expensive, they are very small, for chest images) with low-dose mid-range kV energy than high- very efficient and produce extremely precise x-ray exposures. kV imaging. They do not have special requirements to the mains, do not need synchronisation, and produce Ua independent of Ia. A very impor- tant feature of these high-voltage generators is that the high volt- High-pass filter age can be controlled by controlling the frequency. This is obvious (General) A high-pass filter is a filter that enhances high frequen- from the preceding equation, bearing in mind that for given x-ray cies and reduces low frequencies. The high frequencies models generator, not only the transformer ratio n, but also the size of the high gradient regions in an image while low frequencies model transformer A are constants. This way U is linearly proportional larger structures. Therefore high-pass filters can be used to to f. The converter can smoothly change the frequency, which sharpen edges, e.g. to reveal structures hidden behind a uniformly leads to smooth change of the anode voltage (kV). Additionally distributed source. these generators have much fewer pulses (i.e. less kVp ripple – see the article Voltage waveform). High-resolution CT (HRCT) Related Articles: High-voltage generator, High-voltage trans- (Diagnostic Radiology) High-resolution computed tomography former, Capacitor-discharge generator, Pulse-less generator, (HRCT) is an imaging diagnostic method using precision high- Filament circuit, High-voltage circuit, Voltage waveform resolution computer tomography equipment and dedicated imag- ing techniques which increase the resolution of the output CT High-intensity focused ultrasound (HIFU) image. (Ultrasound) HIFU is a technique whereby pathogenic tissue The scanning protocol used for HRCT ensures image acquisi- is heated and destroyed by high energy focused ultrasound. In tion in thin slices (less or around 1mm slice thickness). order to achieve accurate control of cell death, the transducer HRCT Equipment: HRCT can be performed using a standard is designed to produce highly focussed peak intensities, typi- type of CT equipment which supports narrow slice thickness cally from 1000–2000 W/cm2 in a precisely controlled volume. (around 1mm). Transducer types include shaped single element transducers, a HRCT Data Processing: The data processing algorithms in plane transducer with a lens and multi-element phased arrays. HRCT use image reconstruction techniques that increase the res- Typical frequencies are 0.5–5 MHz. These intensities produce olution of the output image. high temperatures (typically 65°C) in a small contained volume, Other Factors Affecting the Image Resolution: Pixel size, typically 1–2 cm length, 103 mm diameter. If larger volumes focal spot size/dimensions, FOV size/dimensions. of tissue are to be ablated, then the transducer is swept over the Drawbacks of the Diagnostic Method: Smaller field size/ larger volume. Monitoring of the heating is performed using MRI FOV, lower image quality due to thinner slices, higher dose to the or ultrasound imaging. patient and increased x-ray tube load. Trials are under way to use the procedure in the treatment of HRCT Applications in Diagnostic Radiology: HRCT may H cancers including those in prostate, liver, kidney, breast and bone. be generally used for diagnostics of various body areas, the most The technique has been proposed to induce haemostasis in vascu- common being the thorax (lungs) exam. lar trauma. One of the stated advantages of HIFU is that it has no accumulative adverse effect on surrounding tissue. High-voltage cable Further Reading: Ter Haar, G. 2001. Turning up the power: (Diagnostic Radiology) The high-voltage cables are used to lead High intensity focused ultrasound (HIFU) for the treatment of the high-voltage (HV) potential from the HV box (tank) to the cancer. Ultrasound 15: 73–77. x-ray tube. The HV cables have multi-wire copper leads with special rubber or plastic insulation (above 100 kV). A grounded High kV technique metal shielding is placed above the insulation, and over it there (Diagnostic Radiology) High kilovoltage x-ray imaging uses is another protective cover (most often plastic) (Figure H.19). normally potentials higher than 100 kVp. It has special impor- tance for imaging parts of the body with large absorption dif- ferences (e.g. bones and soft tissue). Due to this reason one of the most frequent uses of high-kV technique is in chest radiog- raphy. In this case the decreased image contrast associated with the high energy photons is of lesser importance as the contrast difference between ribs and lungs is very significant. Another area where high-kV x-ray imaging is useful is contrast examina- tions (due to the high absorption of the contrast media – iodine, barium, etc.). One of the advantages of this technique is reducing the patient’s absorbed dose (due to the increased penetration of higher energy photons). Another advantage is the reduced time of the exposure (very important also in the chest region). However in this energy region Compton scattering is the predominant interaction, which requires use of more effective anti scatter grids (high grid ratio). X-ray chest systems use focus to detector distance of the order of 200 cm and in this way there is smaller influence of the x-ray tube focal spot to the overall image resolution. FIGURE H.19 X-ray high-voltage cable (sectioned). Note the metal However, a number of contemporary measurements show that shielding under the surface with protective cotton cover. (Courtesy of the new digital detectors produce better image quality (especially EMERALD project, www .emerald2 .eu) High-voltage circuit 446 High-voltage circuit easily be damaged by the negative wave of HV electricity. This is due to the fact that during this negative wave the already hot anode emits thermo-electrons, which bombard the thin cathode (which is now positively charged) and could easily destroy it. Later the HV has been rectified with one power diode (single- wave rectification), known as single-pulse (or single phase) HVG. Further it has used a bridge-rectifying circuit with four diodes (full-wave rectification), known as two-pulse HVG. Although this HVG is more efficient it still produces 100% pulsations of the high voltage (the kV pulsations are explained in the article Voltage waveforms). Such HVG is used for low power x-ray equipment (e.g. dental). To minimise the high-voltage require- ments of the diodes (and the high-voltage cables insulation) the HV transformer is usually made of two halves (having a 0 poten- tial at its central point) which supply, for example +60 kV and −60 kV to form 120 kV potential difference between the anode and the FIGURE H.20 High-voltage cable with cable end. The cable is with pro- cathode (Figure H.22). This method is widely used for all types of tective plastic cover. (Courtesy of EMERALD project, www .emerald2 .eu) HVG and usually the 0 potential point (which is grounded) is the place where the tube current Ia (mA) is measured. The HVG using three-phase electrical power supply has much There are special requirements for the residual capacity of these larger ‘tripled’ transformer and rectifying circuits (Figure H.23). heavy cables. At both ends the HV cables have special cable-ends (Figure H.20), connecting the HV to the x-ray tube. Silicon paste (or insulation oil) is used to assure the electrical safety of the con- nection. The HV cables have additional wires for cathode fila- ment, anode, rotation, control circuitry, etc. Related Articles: X-ray tube, High-voltage generator High-voltage circuit (Diagnostic Radiology) The two main circuits of the high-voltage U1 generator (HVG) are the high-voltage circuit and the filament cir- H cuit. The high-voltage circuit of the HVG controls the high volt- age (kV) supplied to the x-ray tube and thus controls the x-ray spectrum. In classical HVG this control is achieved primarily by changing the input voltage to the high-voltage transformer (HVT). A different method is used in high frequency generators (changing the frequency) – it will be described in the article on High-frequency generators. The high-voltage circuit is diagrammatically presented as part FIGURE H.22 HVG with bridge-rectifying circuit with four diodes. It produces two-pulse kV waveform. of the high-voltage generator and high-frequency generator elec- trical circuits – see the eponymous articles. The most primitive classical HVG consists of a high-voltage transformer (step-up transformer), which is connected to the sin- gle phase mains and supply the x-ray tube with un-rectified HV (Figure H.21). In this case the x-ray tube (XT) serves also as a diode. This type of HVG is not used any more because the XT can R S

T O Tr U1 FIGURE H.23 Six-pulse HVG. Note that the symmetrical secondary FIGURE H.21 Un-rectified HVG. It produces one-pulse kV waveform winding connection of the HVT is with 12 diodes, but produces 6 pulses like a single-diode HVG. kV waveform. High-voltage control device 447 High-voltage generator Normally the secondary winding of the HVT is with star-con- High-voltage control device nection. The three-phase bridge rectifier of this HVG (with six (Diagnostic Radiology) See Radiographic kV control diodes) provides six pulses in one period, and the amplitude of pulsations is much smaller – approx. 14% pulsations. Due to this High-voltage generator reason these HVG produce less soft radiation and are more effi- (Diagnostic Radiology) All electrical circuits which supply and cient. The six-pulse HVG is used for power equipment (100–120 control the electrical high-voltage energy to the x-ray tube are kW). It can produce the same dose output as one two-pulse HVG, called high-voltage x-ray generator (sometimes also x-ray genera- but for much shorter time. tor or HV generator). The two main circuits of the high-voltage Further there are three-phase HVGs, in which HVT has generator (HVG) are the high voltage circuit and the filament two types of secondary winding connection – star and delta circuit. (Figure H.24). In this case the output pulses are doubled (12-pulsed The first (Roentgen’s) high-voltage generator has used single HVG). This double frequency and 12 diodes bridge-rectifier pro- phase electrical supply and has included just one high-voltage duces kV close to constant potential (approx. 3.4% pulsations). transformer (HVT). Later high-voltage rectifiers and other addi- This makes this HVG very effective, producing very short expo- tional circuitry have been added to the HVT. These have further sure times and suitable for high power output (up to 150 kW). developed to use three phase electrical supply, control equipment, Moreover special high voltage is applied to further reduce the electronics, etc. All these design improvements have followed the pulsations of the HVG. This generator is known as pulse-less and basic principles of the first x-ray generator and have used the nor- will be described in the eponymous article. mal mains frequency of 50/60 Hz and HVT with iron core. Due Related Articles: High-voltage generator, High-voltage trans- to this reason these are known by a common name – conventional former, High-frequency generator, Pulse-less generator, Voltage (or classical) high-voltage x-ray generators (HVG). Contemporary waveform high-voltage generators use electrical frequency above 1 kHz and Further Reading: Karadimov, D. 1978. Roentgen Equipment, are known as high-frequency generators – these will be discussed Technika, Sofia, Bulgaria. in a separate eponymous article. The high-voltage generator’s main parts are autotransformer, high-voltage circuit (including high-voltage transformer, recti- fiers and various command circuits), filament circuit (including filament transformer and various command circuits). These are presented in the diagram of a conventional HVG (Figure H.25). The autotransformer (AT) has two main purposes. The first is the line-voltage compensation – the correction of fluctuations R of the main electrical supply (due to external instabilities or volt- S age drop during the exposure – see the article Voltage drop). The H second purpose of AT is selection of the kV for the exposure (the T peak kV–kVp). The AT consists of a single winding on iron core. O Special graphite slides (roles) are used to adjust the input line voltage and to supply voltage to the high-voltage transformer. The AT transformation ratio N varies between 0.5 and 1.5. The AT is placed either in a separate cabinet or in the control console (where all adjusting/command knobs are). The high-voltage transformer (HVT) is part of the high-voltage FIGURE H.24 Twelve-pulse HVG. Note the asymmetrical secondary circuit. It is a step-up transformer with fixed transformation ratio winding connection of the HVT (star and delta). of the order of 500–600. It is powered by the AT and its purpose Auto transformer HV transformer N var 0.5–1.5 kV selector step-up 1:500 HV rectifier X-ray tube Timer A V AC F Select mA selector Filament transformer step-down 10:1 FIGURE H.25 Block diagram of classical high-voltage x-ray generator. High-voltage protection 448 HIS (Hospital information systems) is to supply the XT with high voltage (normally between 20 and that is about 25% of the size of the traditional generator. All cir- 150 kVp). The secondary winding is usually with a grounded (0 cuits are different and one can control the high-voltage by chang- potential) middle point (it has a 0 potential anyhow). Hence the ing the frequency. For these HVG see the article High-frequency HVT provides two halves of high voltage (for the maximal value generators. of 150 kVp – these are +75 kVp to the anode and −75 kVp to the Related Articles: High-voltage generator, High-voltage trans- cathode), thus the electrical safety requirements to the HV cables former, High-frequency generator, Capacitor-discharge genera- are more easily satisfied. A classical HVG transformer has an iron tor, Pulse-less generator, Monoblock generator, Filament circuit, core and is very massive (for power x-ray equipment it weighs High-voltage circuit, Voltage waveform hundreds of kilograms). There are significant ‘magnetic losses’ in its big laminated iron core, which leads to some delay of the High-voltage protection exposure, as well as distortion of the porches (tails) of the x-ray (Diagnostic Radiology) The high-voltage cables, the x-ray tube pulse. In order to diminish this effect some HVT have special and all parts of the high-voltage generator (HVG) require spe- circuitry for constant pre-magnetising. A voltmeter at the primary cial high-voltage (HV) insulation and protection. The insulation side of HVT is used to monitor the selected kVp (this is possible of the cables must withstand more than the maximal high-voltage due to the fixed ratio of HVT). An ampere meter is connected in tension of the equipment (e.g. 150 kV). The high-voltage trans- the middle point of the secondary side of HVT for measuring the former, the rectifiers, and all other HV parts are immersed in anode current during the exposure. special insulating transformer oil, normally withstanding above The high voltage from the HVT is rectified and then supplied 220 kV/cm. The same HV insulation is used inside the x-ray tube to the x-ray tube. All contemporary x-ray equipment use rectified housing. Some x-ray devices have special system testing the HV HV, which increases the efficiency of transforming the electrical protection and related electrical leakage (due mainly to internal energy to x-ray energy. In the past, vacuum tube diodes have been capacitance). used as rectifiers. At present all rectifiers use solid state compo- Related Articles: High-voltage generator, X-ray tube housing nents (see the article High-voltage circuit for description of the rectifiers). The number and type of connection of the rectifiers define the type of XG (2 pulse, 6 pulse, 12 pulse, etc.). High-voltage transformer The HVT is placed (normally together with the rectifiers and (Diagnostic Radiology) The high-voltage transformer (HVT) is the filament transformer) in a metal HV box (called often HV one of the most important parts of the high-voltage generator. It tank). All elements in the HV box are immersed in special insula- transforms the main voltage to the kV supplying the anode–cath- tion oil (normally more than 220 kV/cm insulation). As the HVT ode high voltage. It is part of the high-voltage circuit. generates significant amount of heat, this oil serves also as cooler. HVT is a step-up transformer with fixed transformation The HV box is grounded and sealed with an air-tight lid. ratio of the order of 500–600. Its primary winding is connected H The high-voltage circuit includes also a timer, which inter- to the mains and its secondary supplies the x-ray tube with high rupts the supply to the HVT after a given period of time, thus voltage (normally between 20 and 150 kVp). The secondary controlling the length of the x-ray exposure (ms). There are vari- winding is usually with a grounded middle point (0 potential). ous designs for timers – usually electro-mechanic ones (old equip- Hence, the HVT provides two halves of high voltage (for the ment) and electronic timers. maximal value of 150 kVp – these are +75 kVp to the anode The filament circuit supplies and controls the filament cur- and −75 kVp to the cathode), thus the electrical safety require- rent through the cathode of the x-ray tube. It includes the filament ments (insolation) are more easily satisfied. The transformer transformer (FT) – a step-down transformer with ratio between is immersed in insulation transformer oil (normally more than 1:10 and 1:20, which generates filament current of the order of 3–5 220 kV/cm insulation). A through the cathode wire. This circuit could also include sub- In the classical high-voltage generator this transformer is circuits that have control of the space charge effect; control of the with iron core and is very massive (for power x-ray equipment it anode temperature influence over cathode wire temperature, etc. weights hundreds of kilograms). There are significant ‘magnetic Often this circuit includes special voltage stabiliser. losses’ in its big laminated iron core, which leads to some delay of The filament circuit controls the input to the FT through vari- exposure, as well as distortion of the porches of the x-ray pulse. In able resistor (this way controlling the filament current, hence the order to diminish this effect some HVTs have special circuitry for anode current – mA). Additionally it can include filament current constant pre-magnetising. adjustment resistors (F-select resistors, set by the service engi- Power x-ray equipment using three-phase mains supply use neer). With these the maximal filament current is limited to a set three-phase high-voltage transformer, which has different types value. These adjustments are made with different kV/mA settings of secondary winding connection. Normally these are star-type using the so called three-point technique to properly select the connections for 6 pulse generator or star-delta connection for 12 necessary resistors (for: low kV@ high mA; high kV@low mA; pulse generator. See the diagrams in the article on Three phase low kV@low mA). generator. Other special types of conventional HVG are capacitor- The new high-frequency high-voltage generators (medium- discharge generators, pulse-less (direct current) generators and frequency high-voltage generators) use high-voltage transformer monoblock generators. These will be discussed separately in with ferrite core. This allows use of higher frequencies, which eponymous articles. minimise the size of the ferrite core down to 25% compared with Although the conventional HVG are still produced and used, iron core. a new type of design was introduced in the 1980s – the high- Related Articles: High-voltage generator, High-frequency frequency high-voltage generator (or more correctly medium- generator, High-voltage circuit frequency high-voltage generator). This HVG uses electronic converters which increase the electrical frequency to several kHz. HIS (Hospital information systems) In this new design the high-voltage transformer has ferrite core (Diagnostic Radiology) See Hospital information systems (HIS) Histogram 449 Hogstrom algorithm Histogram tissues are displayed) then diagnostically significant structures (General) A histogram is a graphical representation of tabular may be difficult to identify by eye. This being the case the pixel data where one axis, typically the x-axis represents a domain intensity histogram can be ‘windowed’ to a certain level and range divided into intervals. The value in each interval is denoted not in the pixel intensity histogram (termed window level [WL] and by the height of the bar but by the area under the bar which in window width [WW] respectively). The pixel values within this turn is determined by the number of observations in that specific range are then displayed across the full grey level display range interval. The total area under all the bars equals the total number of the device displaying the image (as outlined in Figure H.27). of observations made (Figure H.26). All pixels with values below this range are displayed as black, The histogram is considered to be one of the seven quality and all above as white. Windowing may be manual or automatic, control (QC) tools. whereupon the image will be analysed and windowed to ‘humps’ Abbreviation: QC = Quality control. within the pixel intensity histogram. Related Article: Histogram Histogram-based intensity windowing Further Reading: Pisano, E. D. Image processing algorithms (Diagnostic Radiology) A pixel value histogram displays the for digital mammography: A pictorial essay. RadioGraphics 20(5). range and magnitude of pixel values within an image and, as dis- Hyperlinks: Pixel Intensity Histogram Characteristics: played in Figure H.27a, comprises an x-axis of pixel value

and a www .a llabo utcir cuits .com/ techn ical- artic les /i mage- histo gram- y-axis of the number of pixels that have this value. chara cteri stics -mach ine -l earni ng -im age -p roces sing/ If an image contains a wide range of pixel values (a CT image for example, where pixel values [i.e. CT number] for a range of HL-7 (Diagnostic Radiology) HL-7 is an accredited Standards 12 Developing Organisation operating in the healthcare arena which aims to develop international healthcare standards. One of those standards is also called HL-7 and is a protocol for formatting, 10 transmitting and receiving data in a healthcare environment. The HL-7 protocol development was started in 1987 with an aim to 8 create a common ‘language’ allowing healthcare applications to share clinical data with each another. The current version is HL-7 6 version 3 which was published in 2005. The name is derived from the seventh and top layer, the appli- 4 cation layer, of the ISO communications model for Open Systems Interconnection since the protocol is independent of all other, 2 lower, communication layers. H 0 Hogstrom algorithm (Radiotherapy) The Hogstrom algorithm is an implementation FIGURE H.26 Example of a histogram. of a pencil beam calculation for electron beams based on the FIGURE H.27 (a) Pixel intensity histogram of CT image displaying the number of pixels with a certain pixel value, (b) the CT image windowed to display a range of pixel values across the full greyscale range of the image display, (c) the CT image windowed to display the entire range of pixel values across the full greyscale range of the image display and (d) the CT image narrowly windowed to display only two pixel values across the full greyscale range of the image display (hence pixels appear either black or white). (www .dicomlibrary .com) Holmium 450 Holt chamber Fermi–Eyges theory. In this theory, a narrow beam of elec- Holmium trons penetrating a medium is characterised by an increased (General) lateral spread with depth due to multiple scattering processes. The Fermi–Eyges theory assumes small angle scattering, under which conditions the spatial distribution of the pencil beam Symbol Ho is very nearly Gaussian at all depths. This Gaussian distribu- Element category Lanthanides tion can be characterised by the mean square of the scattering Mass number A of stable isotope 165 (100%) angle, mean radius-angle covariance and the mean square of the radius. Atomic number Z 67 A broad beam of electrons can be represented by summing the Atomic weight 164.9303 distribution for many smaller pencil beams (Lillicrap et al. 1975). Electronic configuration 1s2 2s2 2p6 3s2 3p6 3d10 4s2 4p6 5s2 4d10 5p6 4f11 6s2 In this case the dose to a point can be calculated by summing the contributions from all the pencils. Melting point 1734 K A heterogeneous medium, such as a patient, is represented by Boiling point 2993 K a stack of slabs, where heterogeneities along the central axis of Density near room temperature 8790 kg/m3 the pencil beam are extended laterally to a semi-infinite homo- geneous slab. The assumptions of slab geometry and small angle scatter cre- History: Holmium was discovered in 1878 by Marc ate some shortcomings in the approach (Figure H.28). Many of Delafontaine and Jacques-Louis Soret. It is a rare metallic earth these were improved upon following the initial implementation, element with a malleable form and high magnetic moment. but more recently Monte Carlo calculations for electron beam Medical Applications dose distributions have become dominant. Heavy Metal Filters in X-ray Imaging: Holmium is commonly Further Readings: Hogstrom, K., M. Mills and P. Almond. used as a filter in the production of high quality of diagnostic 1981. Electron beam dose calculations. Phys. Med. Biol. x-ray images where contrast agent (usually iodine or barium) has 26:445–459; Hogstrom, K. and R. Steadham. 1996. Proceedings been used. In diagnostic radiology maximum contrast is obtained of the 1996 AAPM Summer School. Advanced Medical when the incoming x-ray energy is close to, but slightly above, the Physics Publishing, Madison, WI; Lillicrap, S., P. Wilson K-edge of the contrast agent in question. For iodine the K-edge is and J. Boag. 1975. Dose distribution in high energy electron 33.17 keV, and for barium the K-edge is 33.45 keV. The K-edge beam distributions from narrow beam data. Phys. Med. Biol. of a standard aluminium filter is 1.6 keV, whereas the K-edge of a 20:30–38. holomium filter is 55.6 keV. Thus a holomium heavy metal filter will transmit a significantly narrower spectrum of photon energies H than an aluminium one. The resultant reduction in the fluence of low energy photons will reduce the patient’s dose, and the reduc- tion in the fluence of high energy photons will increase the image contrast. Medical Lasers: As its laser light can readily be transmitted through optical fibres, holmium has found use as an active com- ponent in some medical (holmium-YAG) lasers. Related Articles: Heavy-metal filter, Contrast media, K-edge metal flitter Holotomography (Diagnostic Radiology) Holotomography is a new imaging method used to examine microscopic biological three-dimen- sional objects. The method uses laser source, which illumi- nates the object (e.g. a cell) from several angles. The measured Refractive Index of the sample (from different angles) is used for back-projecting the image. The method is used in cell biology, where cell organelles can be imaged without labelling agents. The resolution of the method is of the order 100 nm. Holt chamber (Radiotherapy) The Plane-Parallel plate ionisation chamber was developed in 1979 by G. Garrett Holt at Memorial Sloan Kettering Cancer Center, New York, to closely fulfil the requirements of a Bragg-Gray detector in electron beams, especially at low energy – FIGURE H.28 Generalised electron pencil beam approach. The irregu- i.e. to produce a negligible disturbance to electron fluence present lar field is decomposed into elements. A ray trace is performed to a point in the irradiated medium. The chamber was built from polysty- of interest assuming slab geometry. The dose at a point is calculated from rene and graphite. a summation of the pencil beams. Related Article: Parallel plate ionisation chamber Homogeneity 451 Hormesis Homogeneity Abbreviations: HC = Homogeneity coefficient and HVL = (Magnetic Resonance) In MRI, the homogeneity of the main mag- Half value layer. netic field governs the consistency of resonant frequency among Related Article: Half value layer (HVL) the protons in a sample, such that good homogeneity is essential Further Reading: IAEA. 2007. Dosimetry in diagnostic radi- if slice selection, phase encoding and frequency encoding are to ology, an international code of practice. TRS 457, International provide correct information from the patient. Atomic Energy Agency, Vienna, Austria. One factor which affects magnetic field homogeneity is mag- net design: solenoidal magnets with large bores generally have Homogeneity index better uniformity than short bore magnets or open systems. In a (Radiotherapy) The homogeneity index (HI) is a mathematical given system, inhomogeneities in the static magnetic field may be tool designed to analyse the uniformity of dose distributions in caused by imperfect coil winding (variations in current densities the target volume. Various formulae have been described in the within the wire), the presence of metal in the local environment or literature for its calculation, e.g. the ratio of the dose received by even the patient themselves. The process of making the magnetic 95% of the planning target volume (D≥95%) to the dose received by field homogeneous is known as shimming. 5% (D≥5%) of the PTV volume. So that the measure is independent of field strength, homo- Related Article: Dose homogeneity geneity is usually expressed in parts per million (p.p.m) relative to the main magnetic field. The level of homogeneity required for routine spin echo imaging is approximately 10 p.p.m. but for Homogeneous field spectroscopy (where we are concerned with tiny differences in (Magnetic Resonance) The two main types of field characterising resonant frequency between metabolites) the minimum homoge- an MRI system are the main static magnetic field (B0) and the RF neity increases by a factor of 100–0.1 p.p.m. field (B1). Both fields need to be homogeneous to provide a perfect Related Articles: Shimming, Image homogeneity, RF homo- MR image, but this is more important for the main magnetic field. geneity, B0 homogeneity The B0 magnet is made as homogeneous as possible, especially at the isocentre, to ensure that no unnecessary artefacts occur. Homogeneity The RF field homogeneity will vary with the interaction (Radiotherapy) Homogeneity in radiotherapy usually refers to the between the RF field and the object being imaged and the choice degree of variation of dose within a given volume. Radiotherapy of RF coil used in the system. Volume coils, such as the bird cage treatment planning and delivery usually strives to produce and coil, produce high B1 homogeneity over most of the coil volume, deliver a uniform dose to the target volume, in other words the giving excellent image uniformity. Surface coils are characterised homogeneous dose to the target. Recognising that it is not always by their inhomogeneity, producing a high SNR at the surface of practical to get full homogeneity of dose in a target, ICRU (1993) the object which rapidly decreases with depth. has recommended that the degree of homogeneity should be kept Abbreviation: RF = Radiofrequency. H within +7% and −5% of the prescribed dose. Related Articles: B0 Homogeneity, B1 Homogeneity Abbreviation: ICRU = International Commission on Radiation Units and Measurements. Hormesis Further Reading: ICRU. 1993. Prescribing, recording, and (Radiation Protection) The current internationally accepted reporting photon beam therapy. Report 50, International Commission framework for radiation protection is based on a model of poten- on Radiation Units and Measurements, Washington, DC. tial harm from exposure to ionising radiation called the Linear No-Threshold Model. This model suggests that at any level of Homogeneity coefficient (HC) received radiation dose, harm may be caused – i.e. a cancer may (Radiotherapy) Homogeneity coefficient, h, is the ratio of the first be induced. The risk of harm (stochastic effects) is proportionate half value layer (HVL1) to the second half value layer (HVL2): to the dose received (i.e. linear with dose). This is described in Figure H.29. HVL However, more recently published evidence would suggest h = 1 HVL that this model is too simplistic and that for a number of reasons 2 low-level exposure to ionising radiation may actually be benefi- The first half value layer (HVL1) is defined as the thickness of cial. This hormetic effect (hormesis) appears in epidemiological the specified material (absorber), which attenuates the air kerma or air kerma rate in the beam to one half of its original value, measured without any absorber. The contribution of all scattered radiation, other than any which might be present initially in the beam, is deemed to be excluded. The second half value layer (HVL2) is equal to the difference between the thickness of an absorber necessary to reduce the air kerma or air kerma rate to one quarter, d1/4, and the value of HVL1: HVL2 = d1/4 - HVL1 Linear no threshold (LNT) model The value of h gives a certain indication about the width of the x-ray spectrum. Its value lies between 0 and 1 with higher values Dose indicating a narrower spectrum. Typical values of h for beams used in diagnostic radiology are between 0.7 and 0.9. FIGURE H.29 The linear no-threshold model. Probability of effect Hospital Information Systems (HIS) 452 H SL (Hue saturation luminance) such as HL-7 as promoted by the IHE (Integrating the Healthcare Enterprise). Hot and cold spots (Radiotherapy) Generally when planning is done for a treatment an irradiation technique is developed which provides a maximum and uniform dose to the planning target volume and minimises the dose to both the treated volume and the irradiated volume as defined by the International Commission on Radiation Units and Measurements (ICRU 50, 1993). However for many reasons Dose in practice it is difficult to achieve the ideal dose distribution and some dose heterogeneity has to be accepted. ICRU 50 (1993) rec- FIGURE H.30 The hormetic model of dose response. ommends that this dose variation should be kept within +7% (hot spot) and −5% (cold spot) of the prescribed dose. Cold spots are of concern inside the planning target volume (PTV) and hot spots in evidence from mortality rates amongst British radiologists, survi- those regions where organs at risk

are located. vors of Chernobyl, and studies in Taiwan. Figure H.30 describes Therefore with these limits in mind the planning objective the possible consequence of hormesis to our understanding of the would be to ensure that the dose anywhere within the PTV does dose-response curve for stochastic effects. not fall below 95% of the prescribed dose. A hot spot represents Possible reasons for such a response have been postulated. a dose outside the PTV that receives a dose larger than 100% of Firstly, it is recognised that life on earth has evolved whilst contin- the specified PTV dose. Additionally, dose volume is also an issue uously exposed to ionising radiation for natural sources, both ter- and it is considered clinically meaningful if the hot spot exceeds restrial and from space. Secondly, it is recognised that the genome 15 mm in volume. of humans and other animals seem to include ‘anti-cancer’ genes Further Readings: ICRU 50. 1993. Prescribing, reporting – genes that if ‘switched on’ appear to protect from the effects of and recording photon beam therapy. Report 50, International being exposed to carcinogens. Ionising radiation exposure at low Commission on Radiation Units and Measurements, Washington, levels seems to be a trigger to switch on these anti-cancer genes. DC; Williams, J. R. and D. Thwaites. 2000. Radiotherapy Physics The International Commission for Radiological Protection in Practice, Oxford Medical Publications, Oxford, UK. has stated in introducing its latest (2007) recommendations that although it is accepted that hormesis is probably real, such a dose- Hot spot response curve could not be used as the basis for a framework (Radiotherapy) See Hot and cold spots H for occupational, medical, or public radiation protection because of the difficulties in specifying where the boundary lies between Hounsfield number radiation doses that are beneficial and those higher doses that are (Diagnostic Radiology) See CT number deleterious. Related Articles: Linear no-threshold model, Radiobiological Hounsfield scale models, Stochastic effects (Diagnostic Radiology) Many CT scanners use the term Further Readings: Berrington, A., et al. 2001. 100 years of Hounsfield units (HU) to express CT numbers (values propor- observations on British radiologists: Mortality from cancer and tional to the linear attenuation coefficient of the material). The other causes 1897–1997. Br. J. Radiol. 74: 507–519; Chen, W. L. et standard scale of HU (Hounsfield scale) ranges from approxi- al. 2004. Is chronic radiation an effective prophylaxis against can- mately -1000 to +3000 HU. cer? J. Am. Phys. Surgeons 9(1): Spring 2004. ICRP Publication Historically, the range of the Hounsfield scale has been based 103, 2007. The 2007 Recommendations of the International on the early CT scanners, where the difference between CT num- Commission on Radiological Protection. bers of air and water has been accepted as 1000, while the most absorbent bones are usually up to three times this absorption dif- Hospital Information Systems (HIS) ference. Thus, the Hounsfield scale (CT number scale) is from (Diagnostic Radiology) A Hospital Information System (HIS) is –1000 (the accepted HU value for air) through 0 (the accepted HU an information system product specifically for use in a medical or value for water) to +3000 (accepted to represent bone with high healthcare environment. It may be in the form of one single inte- absorption). This equals to 4000 CT numbers. In fact, the range grated software product or be composed of a number of separate has been taken as 4096, as this is 212) – i.e. 12 bits have been used systems that function together to manage information related to for the image with HU scale (grey levels) the remaining 4 bits the general administrative, financial and clinical work and service (from 16 bits to 2 bytes) are used for overlaying alphanumerical or delivery of a hospital or healthcare organisation. graphic information. The HIS may include more functionally specific systems such Related Articles: CT number, Grey value, Matrix size as a radiology information system (RIS) or laboratory information system (LIS), or it may share data with, and receive date from, Housing external implementations of such systems. The interconnectivity (Diagnostic Radiology) See X-ray tube housing and sharing of information means that patient and event related data need only be entered once so minimising issues with data HRCT (High-resolution CT) entry errors, mis-registration or duplication of patients or events. (Diagnostic Radiology) See High-resolution CT (HRCT) The sharing of data requires the various independent infor- mation systems, often developed by different organisations, to HSL (Hue saturation luminance) implement recognised information communication standards (General) See Hue saturation luminance (HSL) Probability of effect HSV (Hue saturation value) 453 Human visual response function HSV (Hue saturation value) Further Readings: http: / /en. wikip edia. org /w iki /H SL _ an d (General) See Hue saturation luminance (HSL) _HSV ; Tabakov, S. 2013. Introduction to vision, colour models and image compression. J. Med. Phys. Intern. 1:50–55. HU calibration (Diagnostic Radiology) The calibration relating pixel values in Human visual response function Hounsfield units (HU) to electron density (ED) is an important (General) The human eye contains both rod and cone cells, factor in CT densitometry, especially necessary for dose calcula- photoreceptors that convert light photons to electrical nerve tions in radiotherapy. impulses. Rod photoreceptors are extremely sensitive to light, HU calibration uses special test objects (phantoms) including functioning at lower luminance levels to facilitate vision in inserts with specific electron densities. Usually different CT scan- dark environments. Cones function at higher luminance levels ners have different calibration curves (HU vs. ED). and facilitate colour vision. The variation in the mechanism Related Article: Stoichiometric calibration by which the eye forms images with differing luminance lev- els leads to a non-linear visual response. This being the case Hue the eye is more sensitive to luminance changes at increased (Diagnostic Radiology) See HSL (Hue, saturation, luminance) luminance levels. This has direct implications in the display of medical images, where observers will be naturally more pro- Hue, saturation, luminance (HSL) ficient in identifying relative changes in luminance in bright (General) The well-known RGB (red, green, blue) colour system areas of an image than in dark areas of an image. Furthermore, is very useful for colour image display on a monitor. This additive an observer’s ability to identify luminance changes will depend system (all colours are formed as the sum of various amounts of on the size of the image studied. the three basic RGB colours) is used in devices which emit light. The Barten model acts as a physical model to describe a human The alternative to RGB – is the subtractive CMYK system (using observer’s response to an image of certain luminance and display the three subtractive colours – cyan, magenta, yellow – plus black) size. The model describes the contrast sensitivity function (CSF), is used in devices reflecting colour – as in colour printing on white the reciprocal of the threshold modulation (Mt). The threshold paper. modulation is described as the ratio of the minimum luminance The RGB system has several limitations in digital imaging amplitude to the mean luminance amplitude of a sinusoidal grat- for several reasons. The non-linear distances between the coordi- ing such that the grating has a 50% probability to be detected (as nates (colours), makes less efficient image compression. A system determined from repeated trials with human observers). The CSF which overcomes this problem and creates a result better suited is described as: for the human visual system is the non-linear colour system of HSL (hue, saturation, luminance). In this system the colour space 1 M is created by: hue (wavelength); saturation (amount of colour, CSF = = opt M H t 2j k int what can be expressed as hue + grey); luminance (intensity or XYT brightness). Each one of these three can be distinguished sepa- rately by the human visual system. The fact that the HSL system has one channel for brightness and is a function of spatial frequency, u (in cycles per degree) and and two channels for chrominance (colour) presents a convenient L, the mean luminance (in candelas per meter squared [cd/m2]) of situation, as human vision is very sensitive to brightness and less the grating in question. sensitive to chrominance. This way the brightness channel can Mopt is the optical modulation transfer function (MTF) of be transmitted fully (for example with 8 bits), and the other two the eye and describes how an image is distorted by the pupil and channels can be represented with 4 bits each, what makes 16 bits optical lens and includes effects of spherical aberration and pupil in total – i.e. saves 8 bits from the 24 bits colour and helps the diameter (which itself is a function of luminance due to pupil compression of the image data. The HSL system has been devel- dilation). oped by A. R. Smith. Sometimes it is also called the HSV system jint is spectral noise density and is a combination of pho- (hue, saturation and volume). ton noise and noise generated by the neural system. The effects An alternative colour system is YUV, where Y is the luma of pupil diameter, quantum detection efficiency of the eye and (Luminance) and U and V are the chrominance values. Closely the lateral inhibition surrounding an excited retina cell are related to this system is YCbCr (where Y is the Luminance). included. These colour systems have special relationships with the RGB X, Y and T are spatial and temporal dimensions of the dis- colours. For example, the Y value in the YCbCr includes a sub- played image. For static images, T is the integration time of the stantial green component (as the human eye is most sensitive to eye (approximately 0.1 seconds). X and Y are limited by the maxi- green) and is given with the expression: mum size over which the eye can integrate information. k is a constant. Y = 0.299 R + 0.587 G + 0.114 B Related Articles: DICOM (Digital imaging and communi- cations in medicine), Image display, Greyscale standard display while the other 2 channels have less green respectively more blue function (GSDF), Medical image display, Barten model or red: Further Readings: Barten, P. G. J. 1992. Physical model for the contrast sensitivity of the human eye. Proc. SPIE 1666, Human Cb = −0.1687 R – 0.3313 G + 0.5 B + 128 Vision, Visual Processing, and Digital Display III (27 August); Cr = 0.5 R − 0.4187 G − 0.0813 B + 128 doi: 10.1117/12.135956; National Electrical Manufacturers All these systems are very useful for digital image compressing. Association. 2011. Digital imaging and communications in medi- Related Articles: RGB (red blue green), Lossless compres- cine (DICOM) part 14: Grayscale standard display function. sion, Image compression NEMA; Ramponi, G. and A. Badano. 2017. Method for adapting Humidity correction factor 454 Hydrophone the grayscale standard display function to the aging eye. J. Digit HVL (half value layer) Imaging 30:17–25, doi: 10.1007/s10278-016-9900-2; Reiser, I. (Diagnostic Radiology) See Half value layer (HVL) 2014. Diagnostic radiology physics: A handbook for teachers and students, Chapter 18, IAEA. Hydrogen Hyperlink: Details of the Barten Model: http: / /dic om .ne ma .or (General) g /med ical/ dicom /2017 e /out put /c html/ part1 4 /sec t _A .2 .html Humidity correction factor Symbol H (Radiotherapy) The calibration factor for an ionisation cham- Element category Non-metal ber is valid only for the reference conditions which apply to the Mass number A 1 calibration and are specified by the standard dosimetry labora- Atomic number Z 1 tory. Any departure from the reference conditions when using the Atomic weight 1.008 g/mol ionisation chamber in the user beam should be corrected by using Electronic configuration 1s1 appropriate factors. The reading of a dosimeter is usually cor- Melting point 14.01 K rected for pressure, temperature, humidity, polarity effect and ion recombination. Boiling point 20.28 K The values of W/e, where W is the mean energy required to Density near room temperature 0.0899 g/L create an ion pair in air and e is the electron charge, and the value of the stopping powers appearing in equation used to calculate the absorbed dose from ionisation measurements apply to a dry History: Molecular hydrogen gas was first produced artifi- air. Therefore the ionisation measurements should be done with cially by T. Von Hohenheim in the sixteenth century. It was only a very low

relative humidity. The value of W decreases with the identified as a new element in experiments performed by Henry humidity and therefore at low humidity conditions the amount Cavendish between 1766 and 1781. The name hydrogen is derived of ionisation increases. On the other side the stopping power from hydro, water, and genesis, to make, and was given to the ele- increases with humidity tending to decrease the charge collected ment because of the water formed when hydrogen gas is burned. by the ionisation chamber. The two opposite effects combine so Isotopes of Hydrogen: By far the most common isotope of that the correction factor for humidity is approximately constant hydrogen is protium, 1H, which is stable and forms 99.98% of nat- for the range of humidities normally encountered by the user at urally occurring hydrogen. Hydrogen also exists in trace amounts room temperature. as stable 2H, deuterium, and radioactive tritium, 3H. As a whole, A humidity calibration factor of 0.997 should be applied for hydrogen makes up over 75% of the universe by mass. There also an ionisation chamber whose calibration factor refers to dry air, exist unstable isotopes of hydrogen from 4H to 7H, which do not H usually for calibrations in 60Co in terms of air kerma. The factor is occur naturally. Hydrogen is formed as a by-product of splitting constant for a large variety of electron and photon beam qualities hydrocarbons. It can also be produced by electrolysis at high cost. used in radiotherapy. Tritium decays to 3He through low energy beta particle emis- No corrections for humidity are needed if the calibration fac- sion, with a half-life of 12.32 years. tor refers to a relative humidity of 50% and is used in a relative Medical Applications: NMR – 1H is the source of protons humidity between 20% and 80%. detected in most NMR investigations. Signal from the proton of A high relative humidity can cause higher leakage currents 1H bound in biological tissue can be collected to obtain infor- in the ionisation chamber and in the electrometer. Therefore it mation about the chemical composition of the tissue, as in NMR is advisable to check them mainly when the ionisation cham- spectroscopy, or to form an image of the hydrogen distribution ber is built from hygroscopic materials such as A-150 plastic and environment, as in MRI. or nylon. Deuterated water (2H2O), or heavy water, is used instead of protium-based water as a solvent for NMR spectroscopy phan- Huygens’ principle toms. Normal water protons would interfere with the signal from (Ultrasound) See Diffraction the chemical solutes used to obtain the spectrum. The 2H proton produces a signal at a different frequency, enabling it to be sepa- Hybrid computation phantoms rated from the solute signals. (Diagnostic Radiology) Computational anatomical phantoms Biomedical Research Tracer – Tritium, 3H, is a common (software phantoms) are used often in radiotherapy (also in radia- marker used to investigate physiological processes, particularly tion protection and imaging) for estimation of organ doses. These in biomedical research and less commonly as a tracer in clinical are mainly Stylised (or mathematical) and Voxel (or tomographic) PET studies. Since hydrogen is found in most biological mole- phantoms. The Stylised phantoms are an earlier type of such cules, 3H can be bound to a wide range of metabolites and tissues. phantoms. They use geometric surface equations, which describe The electron produced in its decay has an average energy of 5.7 the organs and their surfaces. The newer Voxel phantoms use CT keV, so it cannot penetrate the body. Tritiated water is used as a or MR data to create 3D arrays of voxels. The Hybrid software base for many research products and drug safety studies. phantoms are created in a way to keep the mathematical flexibil- Related Article: Magnetic resonance imaging (MRI) ity of the Stylised phantoms, while using the anatomic realism of the Voxel phantoms. The development of these phantoms uses Hydrophone specific software or MATLAB®. (Ultrasound) A hydrophone is a device used for listening to sound Related Article: Software phantoms in liquids, like a microphone does in air. They are made of a piezo- Further Reading: Bolch, W., C. Lee, M. Wayson and P. electric material that converts the pressure changes in the sound Johnson. 2010. Hybrid computational phantoms for medical dose waves to electrical signals and the sensitivity of a hydrophone is reconstruction. Radiat. Environ. Biophys. 49(2)(May):155–168. expressed in V/Pa. Ideally they should be small, sensitive, linear Hyperechoic 455 Hyperfractionation with broad bandwidth. There are three different commonly used film only overlap in a tiny area (diameter 0.05–0.5 mm), which is types of hydrophones: ceramic hydrophones, pvdf needle-probe poled and forms the active sensor element. The advantage with hydrophones and membrane hydrophones. this type of hydrophone is the stable frequency response. The ceramic hydrophone is constructed like a small single Related Article: Acoustic pressure element transducer with a disc-shaped sensor made of absorbing materials. Because of resonance and reflection in the material the Hyperechoic accuracy will be poor. This can be prevented if the hydrophone is (Ultrasound) See Echogenic properly calibrated. Ceramic hydrophones are robust devices and are often used in physiotherapy equipment. Hyperfractionation The needle-probe hydrophone, Figure H.31, is very similar to (Radiotherapy) Hyperfractionation refers to the use of a high the ceramic hydrophone but is smaller (diameter 0.02–1 mm) and number of fractions, with dose fractions smaller than the conven- uses a piezoelectric polymer film as the sensor material, polyvi- tional 1.8–2 Gy per fraction, usually delivered more than once nylidene fluoride (pvdf). The acoustic impedance of the pvdf-film a day. Where this results in an overall treatment time less than is almost the same as for water and this is important to reduce fre- that for conventional fractionation it is referred to as accelerated quency variations of sensitivity. So the needle-probe hydrophone hyperfractionation. is fairly frequency stable but for lower frequencies a rapid fall-off Generally, radiotherapy treatment is delivered once a day, 5 of sensitivity due to diffraction around the probe tip. This type days a week for up to 8 weeks. Each daily treatment is called of hydrophone is commonly used to characterise ultrasonic fields a fraction and such regimens have been shown to be clinically from diagnostic equipments. effective and acceptable, giving a favourable therapeutic effect in A membrane hydrophone, Figure H.32, is also using pvdf as most cases: better tumour control is obtained for a given level sensor material. The polymer film is stretched over an annular of normal tissue toxicity when the radiation dose is fractionated ring of about 100 mm diameter. The electrodes on each side of the rather than delivered as a single dose. The effectiveness of frac- tionated radiotherapy can be understood by consideration of the 5R’s of radiobiology. Dividing the radiotherapy dose into frac- tions spares normal tissues since the cells can repair some of the radiation damage in the time between fractions and cell repopu- lation will occur provided the overall time is sufficiently long. Conversely, fractionating the dose increases the damage to the tumour due to reoxygenation and the redistribution of cells into radiosensitive phases of the cell cycle between fractions. Although the prolongation of treatment by the use of fraction- ation is beneficial in many cases, the excessive prolongation of H treatment may actually have a detrimental effect on the thera- peutic efficacy of the treatment if proliferation of tumour cells becomes significant. Irradiation of tumour cells can trigger the surviving cells to divide faster than before. This is known as accelerated repopulation and further details can be found in the article on Repopulation. Hyperfractionation, particularly acceler- ated hyperfractionation, offers a possible solution to this problem. The aim of hyperfractionated radiotherapy is to exploit the dif- ference in the effect of the dose per fraction on the tumour and FIGURE H.31 Needle-probe hydrophone. (Courtesy of EMIT project, acute-responding normal tissues compared with late-responding www .emerald2 .eu) normal tissues, thereby decreasing the incidence of late normal tissue effects whilst maintaining both the tumour control and the incidence of early effects of the conventional treatment. In clini- cal practice, this results in an increase in the total delivered dose and sometimes also an increase in the overall treatment time. Accelerated treatment aims to reduce repopulation in rapidly pro- liferating tumours by delivering the same total treatment dose as the conventional regimen but in a shortened overall time achieved by giving two or more fractions a day. In practice, it is not pos- sible to achieve this due to the increased incidence of early nor- mal tissue effects. Accelerated hyperfractionated treatment is a combination of the accelerated and hyperfractionated treatment protocols. A low total dose is delivered in a short overall treat- ment time by the delivery of multiple, small doses given at least 6 h apart. Continuous hyperfractionated accelerated radiation therapy (CHART), which reduces overall treatment from 6 to 7 weeks to 12 days and gives 36 small fractions, has been tested in multicentre randomised controlled clinical trials. The trial in non- small-cell lung cancer showed improvement in survival and this FIGURE H.32 Membrane hydrophone. (Courtesy of EMIT project, regimen is now the government recommended standard of care www .emerald2 .eu) for eligible patients in the United Kingdom. It should be noted Hyperpolarised 456 Hypofractionation that early normal tissue effects were more severe than for a con- Conversely, specialised imaging techniques are needed in view of ventional regime but patients found them tolerable and they did this very slow relaxation. not translate to an increase in late normal tissue toxicity. The main application of hyperpolarised gas imaging to date In cases where a change in fractionation regimen is consid- has been ventilation imaging of the lungs. 129Xe in particular ered, perhaps due to an interruption of treatment or as part of a also has promise as a perfusion agent, since it is highly soluble clinical trial, it is useful to be able to compare the regimens in in blood. terms of the effect on both the tumour and the normal tissues. More recently, hyperpolarisation of 13C has been described, This is possible by using the linear-quadratic model to evaluate utilising NMR polarisation transfer from hydrogen polarised the biological effective dose (BED). In the case of normal tissues, using paramagnetic metal catalysts. The resulting 13C has so far a more useful parameter may be the equivalent total dose in 2 Gy been exploited mainly in angiographic applications, but has tre- fractions (EQD2) since most clinical experience of normal tis- mendous potential for molecular imaging too. sue tolerance has been obtained from 2 Gy per fraction regimens. Related Articles: Boltzmann distribution, Longitudinal mag- More detail can be found in the article on Biological effective netisation, Static field dose. Abbreviations: BED = Biological effective dose and EQD2 = Hyperthermia treatment Equivalent total dose in 2 Gy fractions. (Radiotherapy) Hyperthermia treatment is a non-invasive method Related Articles: Adverse effects, Alpha–beta ratio, Biological of increasing tumour temperature to stimulate blood flow, increas- effective dose, Cell cycle, Fractionation, Interruption of treat- ing oxygenation and rendering tumour cells more sensitive to ment, Linear quadratic (LQ) model, Radiosensitivity, Repair, radiation. By adding hyperthermia to radiation therapy, radiation Repopulation, Reoxygenation, Therapeutic effect, Tolerance, 5R’s oncologists can increase tumour control while minimising dam- of radiobiology age to healthy tissue. Very high temperatures can kill cancer cells Further Readings: Hall, E. J. and A. J. Giaccia. 2006. outright (thermal ablation), but they also can injure or kill normal Radiobiology for the Radiologist, 6th edn., Lippincott Williams & cells and tissues. Wilkins, Philadelphia, PA; Saunders, M. et al. 1999. Continuous, hyperfractionated, accelerated radiotherapy (CHART) versus Hypoechoic conventional radiotherapy in non-small cell lung cancer: Mature (Ultrasound) Hypo meaning ‘low’, and echoic ‘of an echo’, data from the randomised multicentre trial. Radiother. Oncol. refers to an object or area with low echogenicity. Echogenicity 52: 137–148; Steel, G. G. 2002. Basic Clinical Radiobiology, is a measure of the ability of objects to reflect ultrasound waves. 3rd edn., Arnold Publishers, London, UK; The Royal College Reflections are a result of a change in acoustic impedance along of Radiologists. 2006. Radiotherapy Dose-Fractionation, The the ultrasound waves’ propagation path. During clinical imaging Royal College of Radiologists, London, UK. this most commonly occurs at tissue boundaries. H Where there is a visible difference in contrast on an ultra- Hyperpolarised sound image

the hypoechoic areas will be grey and appear darker, (Magnetic Resonance) The size of the longitudinal magnetisa- conversely hyperechoic areas which will be whiter and appear tion available in an NMR experiment places a practical limit brighter. An anechoic area will be black. on MRI spatial and temporal resolution. In most cases, this Hypoechoic is often used clinically to describe a mass, cyst, magnetisation is generated by and evolves in the static field of nodules or lesions (Figure H.33). the MRI magnet. Because the difference in populations of the two energy levels brought about by exposure to the static field is Hypofractionation very small (a few nuclei per million), this magnetisation and the (Radiotherapy) Hypofractionation refers to the use of a lower resulting signal are also correspondingly small. However, there number of fractions, with dose fractions substantially larger than are situations in which much larger magnetisation is created, the conventional 2 Gy per fraction. either by transient exposure to a very strong polarising mag- Hypofractionation and single high-dose irradiation are widely netic field or by other means. Although this magnetisation then used in palliative radiotherapy with doses typically ranging from evolves in a weaker field, far more signal is available because 3 to 10 Gy. Lower total doses are delivered than for radical treat- of the larger initial magnetisation. This technique is known as ment and this combined with the limited life expectancy of the hyperpolarisation. patients means that late normal tissue damage is not a major From the Boltzmann equation, it follows that exposure to a very strong field, combined if possible with cooling to very low temperature, would greatly increase the size of the available mag- netisation. This approach has been exploited in systems that use a transient strong magnetic field to induce polarisation, the result- ing magnetisation then evolving in the Earth’s magnetic field. However, in general, less direct methods of hyperpolarisation have been used, usually as a means to image the gases 3He and 129Xe. These methods involve using a laser to generate a large electron spin polarisation, either in the target atom itself or in some intermediary such as rubidium, which is then transferred to the nuclei of interest via hyperfine interactions within the atom, or via collisions. These techniques can result in close to 100% polarisation of spins within the gas. Once polarised, gases can be stored and transported over considerable distances to an imaging FIGURE H.33 Three cyst targets – Anechoic (left), hypoechoic (centre), facility, since they can have relaxation times of 100 h or more. hyperechoic (right). Hypothesis 457 Hypoxia concern. A number of randomised clinical trials of such regimens Related Articles: Adverse effect, Alpha–beta ratio, Biological have demonstrated equivocal symptom control to more fraction- effective dose, Cell cycle, Fractions, Fractionation, Interruption ated schedules and have the advantage of being more convenient of treatment, Linear quadratic (LQ) model, Normal tissue tox- for the patient. icity, Radiosensitivity, Repair, Repopulation, Reoxygenation, Recently, interest in hypofractionation for radical treatment Therapeutic effect, Tolerance, Tumour control probability, 5R’s has renewed in the light of new data on the alpha–beta ratio val- of radiobiology ues for some tumours, the results of fractionation trials such as Further Readings: Bentzen, S. M. et al. 2008a. The UK START, and the advent of intensity modulated radiation therapy. standardisation of breast radiotherapy (START) trial B of radio- Many of the hypofractionated regimens currently in use for radi- therapy hypofractionation for treatment of early breast cancer: cal radiotherapy have developed as a result of expediency rather A randomised trial. Lancet 29: 1098–1107; Bentzen, S. M. et al. than from radiobiological principles. Historically, the most com- 2008b. The UK standardisation of breast radiotherapy (START) mon dose per fraction in use has been 2 Gy. However, in the trial A of radiotherapy hypofractionation for treatment of early northern United Kingdom fraction sizes of 2.67–2.75 Gy per day breast cancer: A randomised trial. Lancet Oncol. 9: 331–341; Hall, have been used for many years, introduced largely to ease the bur- E. J. and A. J. Giaccia. 2006. Radiobiology for the Radiologist, den on thinly spread resources. 6th edn., Lippincott Williams & Wilkins, Philadelphia, PA; The moderate hypofractionated regimes used routinely in Steel, G. G. 2002. Basic Clinical Radiobiology, 3rd edn., Arnold some centres in the north of the United Kingdom are delivered Publishers, London, UK; The Royal College of Radiologists. with slightly lower total doses applied than those for conventional 2006. Radiotherapy Dose-Fractionation, The Royal College of 2 Gy per fraction regimens (e.g. for carcinoma of the prostate 55 Radiologists, London, UK. Gy in 20 fractions has been compared with 66 Gy in 33 fractions) to reduce the risk of late normal tissue damage. For tumours with Hypothesis a high alpha–beta ratio this may lead to a reduction in tumour (General) A hypothesis is a proposition that has not been ruled control probability, although some of the negative effect may be out or proved by experiments or falsified by contradictions to compensated for by the reduced overall treatment time. However, established laws of physics. A hypothesis can be considered as a for tumours with low alpha–beta ratios, in the range of 2–3 Gy possible explanation to an observed phenomenon. The formula- and therefore similar to that for late responding tissues, delivery tion and testing of a hypothesis is a central part of the scientific of a smaller number of larger dose fractions should result in good method. A scientific hypothesis is an idea yet to be accepted as a local tumour control without an increase in normal tissue toxicity. scientific theory. A scientific theory offers a logically self-consis- In the case of prostate cancer, evidence is accumulating that its tent framework that describes processes and phenomena based on alpha–beta ratio is in the low region. In the United Kingdom, the fundamental principles. CHHIP randomised controlled trial will compare 2 Gy dose per fraction (total dose 74 Gy) and 3 Gy dose per fraction (total dose Hypoxia H 57 and 60 Gy). In breast cancer, the START B trial compared 50 (Radiotherapy) Hypoxia is defined as the inadequate supply of Gy delivered in 20 fractions to 40 Gy delivered in 15 fractions and oxygen (O2) to tissue, affecting normal biological functions. found no difference in local tumour control between the two arms Hypoxia can be caused by a number of factors, including but but lower rates of late adverse effects were observed in the hypo- not limited to: fractionated arm. Analysis of the data from the START A trial indicated an alpha–beta ratio for breast cancer of around 4 Gy. • Low O2 partial pressure in arterial blood Intensity modulated radiation therapy (IMRT) and other • Reduced ability of blood to carry O2 (e.g. in anaemia) highly conformal techniques such as tomotherapy and proton • Reduced tissue perfusion therapy, potentially deliver improved dose distributions over con- • Deterioration of the diffusion geometry, (e.g., due to ventional radiotherapy with lower normal tissue doses for com- chaotic vasculature of tumours) parable tumour doses. This may mean it is possible to increase • Inability of cells to use O2 due to poisoning (e.g. due to the dose per fraction delivered to the tumour since the lower dose cyanide) per fraction delivered to late responding normal tissues reduces the need to spare them. The delivery of a simultaneous integrated Acute Hypoxia (perfusion-limited): Acute hypoxia is a boost (SIB), whereby different doses per fraction are delivered to dynamic form of hypoxia and is transient in nature. It occurs in different target regions, is gaining popularity and is incorporated cells which are dependent on blood vessels that are subject to par- into a number of IMRT clinical trials. tial decrease in functionality. This partial (or temporal) decrease In cases where a change in fractionation regimen is consid- in functionality limits oxygen perfusion to the tissue or tumour. ered, perhaps due to an interruption of treatment or as part of a Chronic Hypoxia (diffusion-limited): Limited diffusion of clinical trial, it is useful to be able to compare the regimens in oxygen into tissue leads to chronic hypoxia. The main reason for terms of the effect on both the tumour and the normal tissues. limitation in oxygen diffusion is the distance from the cells to the This is possible by using the linear-quadratic model to evaluate closest blood vessel. the biological effective dose (BED). In the case of normal tissues, Acute (temporary) and chronic hypoxia have been shown to a more useful parameter may be the equivalent total dose in 2 Gy exist in animal and human tumours. Tumour hypoxia is often fractions (EQD2) since most clinical experience of normal tis- reported in terms of the partial pressure of oxygen (pO2) with sue tolerance has been obtained from 2 Gy per fraction regimens. a threshold of 10 mm Hg. The radiosensitivity of a tumour cell More detail can be found in the article on Biological effective strongly depends on its oxygenation level, so if the blood supply is dose. poor the tumour will contain hypoxic cells and be less responsive Abbreviations: BED = Biological effective dose, EQD2 = to radiotherapy. This may occur both due to a reduced rate of free Equivalent total dose in 2 Gy fractions, IMRT = Intensity modu- radical production as well as the oxygenation fixation hypothesis lated radiation therapy and SIB = Simultaneous integrated boost. of DNA repair. Hypoxia targetting 458 H z (hertz) FIGURE H.34 Illustration of the oxygen fixation hypothesis. Large dots (next to R and to RO2) indicate unpaired electrons of free radicals. The Oxygen ‘Fixation’ Hypothesis: The presence of molecu- (perfusion limited) cells, during the treatment induced tumour lar oxygen results in the DNA damage becoming more perma- shrinkage process. nent or fixed, known as the oxygen effect or the oxygen fixation Further Readings: Ewing, D. 1998. The oxygen fixation hypothesis. hypothesis: A reevaluation. Am. J. Clin. Oncol. 21(4):355–361; Indirect interactions constitute approximately 70% of the total Hall, E. 1994. Radiobiology for the Radiologist, 4th edn., J. P. interactions between ionising radiation and DNA. Figure H.34 Lippincott Company; Hockel, M. and P. Vaupel. 2001. Tumor illustrates such an interaction, where the water molecule plays hypoxia: Definitions and current clinical, biologic, and molecu- the role of a ‘mediator’ through the formation of a hydroxyl radi- lar aspects. J. Nat. Cancer Institute 93(4) (February 21); Nias, H cal which subsequently damages the DNA. As long as the dam- A. 2000. An Introduction to Radiobiology, 2nd edn., John Wiley age consists of a single strand break, it can be easily repaired. & Sons; Puck, T. T. and P. I. Markus. 1956. Action of X-rays on However, in the presence of oxygen, an oxygen molecule interacts mammalian cells. J. Exp. Med. 103:653–666. quickly with a free radical, producing an even more damaging free radical (hydroxy peroxide). The damage to the DNA cre- ated by this active radical (RO2) is permanent, since the reaction Hypoxia targetting results in chemical changes to the DNA, which are non-restor- (Nuclear Medicine) This term refers to radiopharmaceuticals that able. Therefore, in the presence of oxygen, the DNA break is ‘set’, target hypoxic cells. It can be used as an assessment of tumour making it unrepairable. This process is called ‘the oxygen fixation hypoxia prior to or during radiotherapy. With such information it hypothesis’ (here ‘to fix’ = ‘to set’, ‘to make it permanent’). On is possible to select patients in need of treatment with radiosensi- the other hand, hypoxic cells would easily repair the damage to tisers or bioreductive drugs. their DNA, since the damaging reaction described above cannot Related Articles: Tracer kinetic modelling, Receptor target- take place in the absence of oxygen. ing, Antigen targeting, DNA targeting, Glycolysis targeting, Therefore, hypoxic cells are more radioresistant than Apoptosis targeting, Neuroreceptor nor- moxic (normally oxygenated) cells due to their ability to repair Further Reading: Seyed, K. I. 2005. Molecular nuclear radiation-created damage. imaging: The radiopharmaceuticals (review). Cancer Biother. The number of cells affected by hypoxia and the specific Radiopharm. 20(2):163–172. oxygen levels of these cells can change during radiotherapy. An increase in oxygen levels is named reoxygenation and is Hz (hertz) often associated with the decrease in the number of chronically (General) See Hertz (Hz) I IAEA Tabakov, S., P. Sprawls, A. Krisanachinda, E. Podgorsak and C. (General) See International Atomic Energy Agency (IAEA)

Lewis. 2013. IOMP model curriculum for post-graduate (MSc- level) education programme on medical physics. J. Med. Phys. IAEA Guides for Medical Physics Int. 1:16–22. (General) The International Atomic Energy Agency (IAEA) has Hyperlink: www .i aea .o rg /pu blica tions /sear ch /ty pe /tr ainin g always been active in the field of education and training related to -cou rse -s eries the application of radiation in medicine. The IAEA has produced several specific Guides related to education and training. These IAEA Handbooks for Medical Physics publications have been endorsed by the IOMP and are also used (General) The International Atomic Energy Agency (IAEA) has as Guides for IOMP accreditation and other professional activi- always been active in the field of education and training related ties. The Guides are part of the following IAEA Training Courses to the application of radiation in medicine. Since 2000, the IAEA Series (TCS) for medical physicists: has commissioned the preparation of three Handbooks to support the education of young medical physicists (related to Physics of • Clinical Training of Medical Physicists Specializing in Radiotherapy, Physics of Diagnostic Radiology and Physics of Radiation Oncology, TCS 37 Nuclear Medicine). The e-books of these three publications are • Clinical Training of Medical Physicists Specializing in free and are currently used in many MSc level university courses, Diagnostic Radiology, TCS 47 alongside IAEA Guides and the RPOP website. Most of these • Clinical Training of Medical Physicists Specializing in publications have been endorsed by the IOMP. Nuclear Medicine, TCS 50 Related Articles: IAEA Guides for Medical Physics • Postgraduate Medical Physics Academic Programmes, Further Readings: International Atomic Energy Agency. TCS 56 2005. Radiation Oncology Physics: A Handbook for Teachers and Students, IAEA, Vienna – www -p ub .ia ea .or g /boo ks /IA TCS 56 also includes the results of the IOMP Model Curriculum EABoo ks /70 86 /Ra diati on -On colog y -Phy sics- A -Han dbook -for- project. These publications are free and are currently used glob- Teach ers -a nd -St udent s; International Atomic Energy Agency. ally to support the formation of MSc level university courses and 2014. Diagnostic Radiology Physics: A Handbook for Teachers related clinical training. and Students, IAEA, Vienna – www -p ub .ia ea .or g /boo ks /IA Other useful IAEA reports important for the field of medical EABoo ks /88 41 /Di agnos tic -R adiol ogy -P hysic s -A -H andbo ok -fo physics include: ‘Medical Physics Staffing Needs in Diagnostic r -Tea chers -and- Stude nts; International Atomic Energy Agency. I Imaging and Radionuclide Therapy: An Activity Based Approach’. 2015. Nuclear Medicine Physics: A Handbook for Teachers and The publication of Loreti et al. (2015) provides a full list of such Students, IAEA, Vienna –www -p ub .ia ea .or g /boo ks /IA EABoo ks reports. All these IAEA publications form the backbone of medi- /10 368 /N uclea r -Med icine -Phys ics -A -Hand book- for -T eache rs -an d cal physics education, training and professional activities in all -Stu dents ; Loreti, G., H. Delis, B. Healy, J. Izewska, G. L. Poli low- and middle-income countries. and A. Meghzifene. 2015. IAEA education and training activities Related Articles: IAEA Handbooks for Medical Physics in medical physics. J. Med. Phys. Int. 3(2):81–87. Further Readings: International Atomic Energy Agency. 2009. Clinical Training of Medical Physicists Specializing in ICNIRP Radiation Oncology, Training Course Series 37, IAEA, Vienna (Radiation Protection) See International Commission on Non- – www -p ub .ia ea .or g /boo ks /IA EABoo ks /82 22 /Cl inica l -Tra ining Ionising Radiation -of -M edica l -Phy sicis ts -Sp ecial izin g -in -R adiat ion -O ncolo gy; International Atomic Energy Agency. 2010. Clinical Training ICRP (International Commission on Radiological Protection) of Medical Physicists Specializing in Diagnostic Radiology, (Radiation Protection) See International Commission on Training Course Series 47, IAEA, Vienna – www -p ub .ia ea .or g Radiological Protection (ICRP) /boo ks /IA EABoo ks /85 74 /Cl inica l -Tra ining -of -M edica l -Phy sicis ts -Sp ecial izing -in -D iagno stic- Radio logy; International Atomic ICRU Energy Agency. 2011. Clinical Training of Medical Physicists (Radiation Protection) See International Commission on Specializing in Nuclear Medicine, Training Course Series 50, Radiation Units and Measurements IAEA, Vienna – www -p ub .ia ea .or g /boo ks /IA EABoo ks /86 56 /Cl inica l -Tra ining -of -M edica l -Phy sicis ts -Sp ecial iz ing -in -N ICRU reference point uclea r -Med icine ; International Atomic Energy Agency. 2013. (Radiotherapy, Brachytherapy) Postgraduate Medical Physics Academic Programmes, Training ICRU Report 38 – Reference Points for Reporting Doses Course Series 56, IAEA, Vienna – http: / /www -pub. iaea. org /b to Organs at Risk: The ICRU reference points for bladder and ooks/ IAEAB ooks/ 10591 /Post gradu ate -M edica l -Phy sics- Acad e rectum are described in the ICRU Report 38 ‘Dose and Volume mic -P rogra mmes; Loreti, G., H. Delis, B. Healy, J. Izewska, G. Specification for Reporting Intracavitary Gynaecology’. The L. Poli and A. Meghzifene. 2015. IAEA education and train- reference points are defined in relation to the applicator and the ing activities in medical physics. J. Med. Phys. Int. 3(2):81–87; doses to these points are calculated from orthogonal radiographs. 459 ICTP (International Centre for Theoretical Physics) 460 Image artefact A Foley catheter, with the balloon filled with 7 cm3 contrast was organised in ICTP in 1988 by A Benini and L Bertocchi. This fluid, is used to define the bladder reference point. On the lateral activity has become regular and 15 colleges had been organised film this point is defined as the posterior surface of the Foley by 2018, educating over 1000 young medical physicists from over balloon, on the vertical AP axis drawn through the centre of 100 low- and middle-income countries (see article ICTP College the balloon. On the frontal (AP) film, the balloon centre point on Medical Physics). is used. In this period, ICTP took part in the EU projects EMERALD, On the lateral film, the rectal reference point is located on a EMIT and EMITEL. It also hosted several important confer- vertical line drawn from the lower end of the intrauterine source. ences (organised by S Tabakov): 1998 – the first International The point is defined 0.5 cm posterior to the vaginal wall, which is Conference on Medical Physics Training; 2003 – the first visualised, e.g. using a radio-opaque gauze packing. On the fron- International Conference on e-Learning in Medical Physics; 2008 tal (AP) film the reference point is located at the lower end of – the Medical Physics Encyclopaedia Conference. the intrauterine sources or at the midpoint between the vaginal In 2014, ICTP established (together with the University sources. of Trieste) the first international MSc programme in Medical ICRU Report 38 – Reference Points for Specification of an Physics, headed by R Padovani and R Longo. The MSc has strong Intracavitary Application: The ICRU Report 38 recommends support from the Italian Association of Medical Physics. the use of ‘reference volume’ together with the total reference air In 2015, ICTP organised a regular School of Medical Physics kerma for the application, instead of dose to a reference point in a for Radiation Therapy (in the alternating years of the College on region with a high dose gradient. Medical Physics). The School is organised by R Padovani and L Nevertheless, two points called A points, defined in relation Bertocchi. to the applicator as ‘2 cm up and 2 cm out laterally left and right’ Alongside its many activities related to theoretical and applied have been used extensively both for specification and reporting physics, every year ICTP hosts various courses organised by (point A was originally defined in the Manchester system). IAEA (related to radiation physics and medical physics). Reference Points – Reporting for Image-Guided Techniques: Related Articles: College on Medical Physics ICTP Modern recommendations for reporting intracavitary brachyther- Further Readings: Bertocchi, L., A. Benini, F. Milano, R. apy are based on image-guided techniques and dose-volume his- Padovani, P. Sprawls and S. Tabakov. 2014. 50 years ICTP and its togram parameters. But, both the ICRU reference points and point activities in the field of medical physics. J. Med. Phys. Int. 2(2): A are still included for reporting purposes, to relate the newer 410–416. image-guided techniques to the older techniques. Abbreviations: AP = Anterior − posterior and ICRU = IDR (Instantaneous dose rate) International Commission on Radiation Units and Measurements. (Radiation Protection) See Instantaneous dose rate (IDR) Related Articles: Internal reference point, Reference volume, Manchester system IEC Further Readings: Gerbaulet, A., R. Pötter, J.-J. Mazeron (Radiation Protection) See International Electrotechnical and E. van Limbergen, eds. 2002. The GEC ESTRO Handbook Commission (IEC) of Brachytherapy, ESTRO, Brussels, Belgium, available at the I ESTRO web site: www .estro .be; ICRU (International Commission IF (intensification factor) on Radiation Units and Measurements, Inc.). 1985. Dose and vol- (Diagnostic Radiology) See Intensification factor (IF) ume specification for reporting intracavitary therapy in gynecol- ogy, ICRU Report 38, Washington, DC. IFMBE (General) See International Federation for Medical and ICTP (International Centre for Theoretical Physics) Biological Engineering (General) The ICTP (the Abdus Salam International Centre for Theoretical Physics) is an international research institute for phys- ical and mathematical sciences, which operates under a tripar- IGRT (Image-guided radiation therapy) tite agreement between the Italian Government, United Nations (Radiotherapy) See Image-guided radiation therapy (IGRT) Educational, Scientific and Cultural Organization (UNESCO), and the International Atomic Energy Agency (IAEA). ICTP was Image analysis founded in 1964 by Mohammad Abdus Salam, a Nobel Laureate (General) Image analysis is the process by which useful informa- in Physics of Pakistani nationality. The Centre buildings are in tion may be extracted from images, and is commonly performed Trieste, Italy. The mission of the ICTP is: To foster the growth of through digital image processing. advanced studies and research in physical and mathematical sci- Image analysis usually entails isolating the information in ences, especially in support of excellence in developing countries; the image which is likely to be most significant, then identifying To develop high-level scientific programmes, keeping in mind the likely ‘objects’ and categorising them by their intensity, colour, needs of developing countries, and provide an international forum shape, pattern, etc. of scientific contact for scientists from all countries; To conduct research at the highest international standards and maintain a Image artefact conducive environment of scientific inquiry for the entire ICTP (Diagnostic Radiology) In diagnostic imaging, the term ‘artefact’ community. The Centre is an institution that is run by scientists refers to structures in the image that are not a true representation for scientists. of the object. In computed tomography (CT), artefacts are repre- The first medical physics activity in ICTP was organised sented by systematic discrepancies between the CT numbers in in 1982 – an International Conference on the Applications of the reconstructed image and the true attenuation coefficients of Physics to Medicine and Biology (organised by G Alberi). The the object. They can be a result of shortcomings in equipment, first College on Medical Physics (focused on medical imaging) physical principles or may be patient related. Image artefact 461 Image enhancement Examples of artefacts due to equipment malfunction are ring Image compression artefacts. Examples of artefacts due to physical principles are (General) Image compression is a digital image processing tech- beam hardening artefacts and partial volume artefacts, and an nique which attempts to reduce the amount of data needed to rep- example of a patient-related artefact is that due to motion. resent the information in an image. CT artefacts are usually presented in the image as shading, Many compression techniques exist which can reduce the origi- streaks or rings. Some can be prevented by frequent recalibra- nal image pixel values into a much smaller coded block of data. The tions and good quality control procedures to ensure correct techniques may be divided into ‘lossy’ and ‘non-lossy’ processes. operation of equipment. Others can be minimised by selec- Non-lossy compression is able to protect all the original image tion of appropriate scan parameters or use of artefact reduction data and it is possible to exactly reconstruct the original image software. without loss of any information.

This is therefore favoured in Related Articles: Artefact, Beam hardening, Cone beam arte- medical applications where images are used for diagnosis, though fact, Helical artefact, Metal artefact, Motion artefact, Partial vol- the amount of compression available (size reduction) may only be ume effect (artefact), Ring artefact within one order of magnitude. Lossy compression techniques are now common where image Image artefact data need to be transmitted quickly, and some amount of image (Radiotherapy) Imaging in radiotherapy is used in planning the loss can be accepted. Common image storage formats such as patient’s treatment. Often CT and/or MRI images are used. CT JPEG and MPEG include compression, though the amount of provides attenuation information and MRI better delineation of compression and the restriction to lossless compression is usually organs. PET images are also used for planning radiation therapy selectable. Lossy compression techniques may reduce image data treatments. US images are used to plan brachytherapy treatments. size by large factors (1/10–1/100) with only minimal reduction in These imaging methods are prone to artefacts. These artefacts observed image quality. may impact on treatment planning, in terms of volume delineation and dose calculation. Image covariance CT Artefacts: In the case of CT the presence of high Z implants (General) A mathematical measure of the similarity of two digital such as artificial hips or implanted gold fiducial markers lead to images. streaking (see Figure I.1). The effects of patient motion, such as that due to breathing, may lead to distortion of the target shape and organ shapes if the CT scanner is not synchronised with the Image display breathing motion using 4D CT. (General) In medical imaging, digital images are displayed on MRI Artefacts: MRI is also prone to artefacts caused by cathode ray tubes (CRT) or liquid crystal displays (LCDs). patient movement during scanning. In addition to this spatial Individual pixels in an image are displayed with different distortion occurs due to variations in magnetic field strength and brightness levels depending on the pixel value. The images can magnetic susceptibility effects. be displayed in black and white (greyscale) where the levels are Abbreviations: CT = Computed tomography, 4D CT = Four represented by varying degrees of white light or colour where the dimensional computed tomography and MRI = Magnetic reso- levels are assigned to a colour scale consisting of varying degrees nance imaging. of red, green and blue light (RGB). I Further Readings: Bal, M. and L. Spies. 2006. Metal artefact Image display systems are characterised by a number of reduction in CT using tissue-class modeling and adaptive prefil- parameters, examples of which are spatial resolution, spatial dis- tering. Med. Phys. 33:2852–2859; Baldwin, L. N., K. Wachowicz, tortion and contrast resolution. In order to preserve image detail, S. D. Thomas, R. Rivest and B. G. Fallone. 2007. Characterization, the spatial resolution of a display device should be greater than prediction, and correction of geometric distortion in 3 T MR that of the underlying image. images. Med. Phys. 34:388–399; Jiang, S. B. 2006. Radiotherapy Related Articles: Grey levels, RGB of mobile tumors. Semin. Radiat. Oncol. 16:239–248. Image enhancement (Ultrasound) Image enhancement is a general term covering adjustments to an image in order to improve or alter the image and its appearance. In medical imaging it encompasses a range of techniques to improve the detection of features in images. Techniques include • Post-processing, ascribing different grey levels or colours to the final output • Adaptive processing (Figure I.2) • Edge enhancement • Temporal filtering, e.g. persistence Adaptive processing produces smoothing in the ultrasound image of a phantom with more consistent contrast in the cylinder targets. The processing reduces background speckle in the phantom gel material but introduces patterns in the image that are not in the original. FIGURE I.1 CT scan of prostate with implanted markers. Note the Related Articles: Post-processing, Adaptive processing, Edge streak artefacts extending diagonally across the centre. enhancement, Persistence Image fusion 462 Image geometric distortion Distortion Posit ion FIGURE I.3 One form of geometric distortion makes it difficult to see the actual positions of objects within the body. (Courtesy of Sprawls Foundation, www .sprawls .org) FIGURE I.2 Adaptive processing. Distortion Shape Image fusion Image (Nuclear Medicine) Nuclear medicine and PET images suffer greatly from a lack of anatomical detail. In order to provide ana- tomical structure, computer algorithms have been developed to accurately align these images with images taken from CT or MRI. It is very useful to display the anatomical detail (e.g. provided Objects in body by CT) and the physiological detail (e.g. provided by SPECT) on the same image, one overlying the other. This is known as image fusion. FIGURE I.4 Depending on the orientation of objects relative to the Related Article: Registration direction of the x-ray beam or image plane, their actual shapes might not be displayed. (Courtesy of Sprawls Foundation, www .sprawls .org) Image Gently (Diagnostic Radiology) Image Gently is a campaign to raise awareness in the imaging community of the need to adjust Size radiation dose when imaging, especially children. It is coordi- Distortion Size I nated by the Image Gently Alliance, a coalition of health care Size Size organisations dedicated to providing safe, high-quality paedi- Magnification = FRD atric imaging worldwide. The ultimate goal of the Alliance is FOD Image to change practice. The Alliance began as a committee within the Society for Pediatric Radiology (SPR) in late 2006. In 2007, The Society leadership reached out to friends and colleagues in sister societies representing the key members of the imag- ing team, ACR, ASRT and AAPM, to form ‘the Writers Group’. These organizations developed the concept of the Alliance and FOD ORD their representatives developed the campaign in the summer of FRD 2007. The organization has developed a transformative group of campaigns to address digital radiography, fluoroscopy, inter- Focal spot Object Receptor ventional radiology, nuclear medicine, computed tomography, dentistry, cardiac imaging and imaging in the setting of minor head trauma. FIGURE I.5 Because of the geometric magnification associated with the imaging process the size of objects within a body is not accurately Hyperlink: www .imagegently .org/ reproduced in the image. (Courtesy of Sprawls Foundation, www .sprawls .org) Image geometric distortion (Diagnostic Radiology) Several forms of image distortion can occur in projection x-ray imaging and are related to the geometry the fact that the peripheral photo electrons from the of the process. Three types of geometric distortion are illustrated II photocathode are less affected by the accelerating later (Figures I.3 through I.5). electromagnetic field. Usually pincushion distortion Other geometric distortions exist in fluoroscopy. The main is prominent at larger II fields (see the eponymous such image intensifier (II) distortions are as follows: article). • Barrel distortion – an opposite to pincushion distortion. • Pincushion distortion – it is caused by a non-uniform This effect can be due to various factors, one of which is magnification across the image, with magnification pre-compensation of the Pincushion distortion (see the increasing away from the image centre. This is due to eponymous article). Image intensifier 463 Image intensifier • S distortion – it is mainly due to external electromag- equipotential lines of the accelerating electric field form an ‘elec- netic fields. As a result the peripheral photo electrons tric lens’ which focuses precisely the beam of photoelectrons (the from the II photocathode are under the influence of both dashed lines on Figure I.6). Any external electromagnetic fields the accelerating electromagnetic field and the external can distort the ‘electric lens’ and hence the image. Due to this field. This combined influence creates S distortion (see reason the II has to be shielded (hence the need of mu-metal for the image in the article on Vignetting). its envelope). Figure I.7 shows an old II with glass envelope. The front phos- Related Articles: Pincushion distortion, Barrel distortion, phor is clearly seen. The diameter of this II is 23 cm (compare Vignetting with the pencil on the photo). Usually II are made with standard front diameters such as 15, 23, 30, 40 cm (in inches from 6 to 16 Image intensifier in. diameter). Figure I.8 shows the same II with broken entrance (Diagnostic Radiology) The image intensifier (II) is an electronic window. The image clearly shows the cylindrical accelerating vacuum device used to intensify electronically the brightness of electrodes and the small output phosphor at the back (usually with the x-ray image. II is the main detector used in x-ray fluoroscopy. diameter of 2.5 cm). The dose necessary to produce an image with the II is signifi- There are a number of parameters used to assess an image cantly less than the dose necessary to produce a film radiograph. intensifier (II). The most important being Typical block diagram of an II is shown on Figure I.6. The envelope (4) of the II is made by glass (in older II types) or • Total brightness gain = (output light photons)/(input mu-metal (newer II). In the case of mu metal the Input win- x-ray photons) dow (also called entrance window) 1 is made of very thin metal (as 0.2 mm aluminium) which assures 95% transmission of the falling x-ray radiation (already modulated by the absorption of the tissues of the patient). The x-ray quanta passing through 1 hit the input phosphor (2) and produce light with intensity proportional to the x-ray beam intensity. Most of the new II use caesium iodide (often CsI:Na) as input phosphor. This material emits light with spectrum matching closely the sensitivity of the next layer – the photocathode 3 (usually made of antimony and caesium). The light photons from the input phosphor inter- act with the photocathode, which produces photoelectrons. The number of photoelectrons produced is proportional to the num- ber of light photons, hence to the intensity of the x-ray beam. A system of accelerating electrodes (5) focuses and accelerates the photoelectrons to approximately 30 keV. These acceler- ated photoelectrons hit the output phosphor 7 and produce a small but very bright light. Most II use zinc sulphide (often I ZnCdS:Ag) as output phosphor. This bright image from the output phosphor passes through a glass window 8 and through fibre optic or optical system (9) is transferred to the TV cam- era tube (and from the to the diagnostic TV monitor). A thin (approx. 0.2 μm) non-transparent metal anode (6) assures that FIGURE I.7 Image intensifier glass envelope. (Courtesy of EMERALD the bright light from the output phosphor does not light back project, www .emerald2 .eu) the photocathode. Usually the electric potential of the photocathode is 0 V, and the potential of the accelerating electrodes increases reach- ing +30 keV at the anode in front of the output phosphor. The 1 2 3 4 5 6 7 8 9 X-rays Light FIGURE I.8 Broken image intensifier to show the accelerating elec- FIGURE I.6 Image intensifier block diagram. trodes and output window. (Image courtesy of A. Litchev.) Image matrix 464 Image noise Usually this figure is between 1000 and 6000. • Conversion factor = (output phosphor light)/(input screen dose rate) Usually this figure is between 100 and 1000 (cd/m2/μGy/s). The tandem optics (with semitransparent mirror) at the output of the II allows splitting the light beam to two – one to the TV camera and monitor, the other to some recording device as cine camera, spot camera, etc. (see related articles). The automatic brightness control (ABC) allows monitoring the dose rate at the entrance window of the II by inserting a probe which measures the intensity of the light exiting the II in the par- allel light beam formed at the optics level (9). The newest flat panel detectors used in digital radiography can work in fluoroscopic mode as well. It is expected that in future these will replace the image intensifiers. Related Articles: Fluoroscopy, Total brightness gain, FIGURE I.10 Image noise (granularity) of a uniform CT scan of bottle Cinefluoroscopy, Spot camera with water (uniform object). Image matrix (Diagnostic Radiology) See Matrix size In digital systems, noise is often estimated by the standard deviation of the pixel values within certain region of interest. Image noise Figure I.10 gives an example of image noise (the visual granular- (Diagnostic Radiology) Visual noise in an image is a generally ity) in a CT scan of bottle with water, which should represent a undesirable characteristic of the image that reduces visibility of uniform

image with CT number 0 (in the image is 0.4), but in real- low contrast objects (Figure I.9). In x-ray and radionuclide imag- ity, the water is presented with a granular structure of different ing the statistically random distribution of photons received by pixel values (St. Dev. 6.62). This representation of the noise in a the receptor is the predominant source of noise. CT image is related to various factors, which for CT are described In general image noise is the stochastic variation of the image by the formula of Brooks and Di Chiro: signal. Quantitatively it is described by the SNR (signal to noise ratio) or NPS (noise power spectrum) of the system. 1/2 There are various sources of noise – the imaging radiation, the é (e-md ) ù detector and system hardware, the reconstruction software, etc. A s ~ C ê ú ê ë (D * ST * P3 ) ú most often encountered noise in imaging systems using ionising û I radiation is quantum noise, which is due to the stochastic varia- tion of x-rays, photons, etc. which shows that the CT noise (σ) depends mainly on • Reconstruction algorithm and filter – C • Object parameters – (μd) Effect of noise on visibility • Scan dose – D Low • Slice thickness – ST • Pixel size – P Related Articles: Contrast resolution, Signal to noise ratio, Noise power spectrum Image noise (Magnetic Resonance) Image noise is unavoidable in MRI, as in all other imaging modalities. There are two main sources of noise in an MR image: electromagnetic noise in the body of the patient due to the movement of charged particles within human tissue, and Johnson-Nyquist noise in the RF coil and associated electron- ics within the MRI system. Under most circumstances, patient- High generated noise is dominant. However, when imaging at low field, Increasing noise or when using small receiver coils even at 3 T, noise generated in the coil can be more significant. A number of parameters affect the amount of noise within an FIGURE I.9 Image noise and object contrast detectability. (Courtesy of image including bandwidth, matrix size, slice width and acquisi- Sprawls Foundation, www .sprawls .org) tion timing parameters such as TR, TE and TI. The choice of RF Object contrast Image noise 465 Image processing software coil(s) will also have an impact, with a small tightly coupled coil two or three physical gradients images can be collected in any generally demonstrating less noise than a larger coil. The amount oblique plane. This extremely useful capability allows images that this contributes to the noise depends on the size of the RF to be collected, for example in the long and short axes of the coil and the bandwidth of the pulse sequence. A large RF coil or heart, which are not conveniently aligned with either ana- a wide bandwidth will increase noise. tomical planes in the patient or the Cartesian axes of the MRI Noise is normally quantified by measuring the signal to noise scanner. ratio (SNR). This is a more useful number than considering noise Related Articles: B0 gradients, Multiplanar reconstruction, alone, since the strength of the signal will determine how much Oblique imaging, Slice selection noise will be acceptable in the image before it ceases to be diag- nostically useful. Random noise gives the MR image a slightly Image processing software mottled effect and the lower the SNR, the more mottled the image (General) Medical imaging systems include built-in image pro- becomes. cessing capability provided by the modality vendor. This capa- As well as noise generated within the imaging process itself, bility usually resides on the acquisition console of the modality MR images are susceptible to external RF noise from a range of or on an associated satellite computer system. Reporting sta- sources, such as radio and television broadcast, noisy electronic tions on PACS systems will also have image processing capa- components in other pieces of equipment, and static electrical bilities. Image processing applications running on clinical discharge. These may also contribute random noise, but in some systems for day to day use guide the user via graphical user cases the noise may be structured, for example the ‘herringbone’ interfaces (GUIs) and do not require any programming skills. artefact due to electrical spikes and discrete bands of noise due to For implementation of more sophisticated techniques or for the narrowband RF transmissions. development of new techniques, off line image processing using Related Articles: Johnson-Nyquist noise dedicated software may be necessary. Several image processing Abbreviations: SNR = Signal to noise ratio. suites appropriate for medical use are available from companies and institutional organisations (e.g. the ImageJ platform from Image noise the National Institutes of Health). Development of new image (Nuclear Medicine) Image noise is a common feature in diag- processing techniques may require programming using lan- nostic imaging, particularly in nuclear medicine where the count guages such as JAVA, C or proprietary languages designed for rates (or information densities) are low. Noise can be divided into use with specific platforms. two categories depending on their nature and behaviour. The first Basic medical imaging processing software functions include category is called random noise, which is basically statistical the following: noise present in all images as a mottle background. Random noise is caused by several factors, e.g. statistical variations in counting • Reformatting: Planar reformatting allows presentation rate and static noise in circuitry. of images in planes other than those originally acquired The second type is structured noise and it refers to nonran- on the modality. For example, a stack of acquired trans- dom effects on the counting rate. Such non-random variations can verse slices could be reformatted to present oblique be caused by uptake in superficial organs blocking the view of a slices through the field of view. Reformatting also deeper lying organ of interest. Other kinds of structural noise are allows acquired slices to be presented with improved I image system artefacts, e.g. non-uniformities in gamma camera SNR (but with reduced resolution). or image reconstruction artefacts. • Filtering: Smoothing filters can be used to reduce the Further Reading: Cherry, S. R., J. A. Sorenson and M. E. ‘grainy’ appearance of images. For example, SPECT Phelps. 2012. Physics in Nuclear Medicine, 4th edn., Elsevier images of the heart may be smoothed by image process- Saunders, Philadelphia, PA, pp. 243–247. ing software prior to presentation to improve conspicu- ity of under perfused myocardium. High pass filters can be used to sharpen edges. Digital mammography sys- Image perception tems may include filtering options to help highlight the (General) Image perception is the process by which the informa- presence of microcalcifications. tion in images are viewed and interpreted. In medical imaging, • Image maths: Pixel by pixel addition, subtraction, mul- the subject encompasses the techniques of image acquisition, pro- tiplication or division of two images to generate a third cessing and display and also human perceptual and interpretation image. Image subtraction is the most commonly used. processes. For example, in vascular imaging a pre contrast injec- tion image ‘mask’ is subtracted from a post-contrast Image plane image to suppress background and highlight the con- (Magnetic Resonance) MRI is inherently a three-dimensional trast filled vasculature. imaging technique, with the additional step of slice selection • Segmentation: Segmentation attempts to separate needed to restrict data acquisition to a slice rather than a volume out specific tissue types in the acquired images. This of tissue. Nevertheless, in most applications images are collected type of processing is common in CT, where segmen- from a stack of slices. The orientation of these slices is known as tation algorithms based on pixel by pixel analysis of the image plane. Hounsfield units can split images into bone, vascular In MRI the image plane is determined by the orientation of and soft tissue components. the slice selection gradient. It can lie in any one of the principal • 3D processing: 2D stacks of images can be reformat- anatomical planes (axial, coronal or sagittal), or by combining ted for presentation as 3D images. This process may be Image quality 466 Image quality indicators combined with segmentation to provide a 3D view of a by using a detection coil optimised to the part of the body particular structure, e.g. the coronary vessels in a CT being imaged, to prevent the detection of noise from adjacent scan. structures, by summing several signal measurements, by sam- • Quantification: All modalities allow some degree of volu- pling larger volumes (reducing spatial resolution by increasing metric and dimensional measurement and tumour dimen- the FOV and slice thickness) and by optimising the receiver sions and stroke volumes may be calculated using simple bandwidth. delineation of structures. Other quantitative data that can Image Contrast and Contrast to Noise Ratio: The contrast be extracted from images by processing are modality and in an MR image depends on the type of pulse sequence used and exam specific. For example, in nuclear medicine image its parameters. Structures with little contrast between them are processing may be used to provide measures of isotope difficult to distinguish in the presence of noise, and this can be uptake, cardiac CT image processing allows calcium expressed in the contrast to noise ratio (CNR), the ratio of the scoring of plaque, brain activation maps can be generated signal difference between two objects to the noise present in the in fMRI and blood flow measures can be extracted from image. MRI phase contrast angiography images. Artefacts: Artefacts are features of the image that are not found in the imaged object, but are introduced due to imperfec- Image quality tions in the imaging process or to confounding factors such as (Magnetic Resonance) patient movement. It is important to be familiar with the appear- Definition: Image quality is a general term used to describe ance of artefacts because they can be mistaken for pathology. how faithfully an image reproduces features of the imaged object Numerous kinds of artefacts can occur in MRI and a few of them and the degree of visibility of relevant information in an image. are summarised in Table I.1. Good image quality is essential to achieve diagnostic quality in Further Readings: Haacke, E. M., R. W. Brown, M. R. MRI and other imaging modalities. It is often assessed using Thompson and R. Venkatesan. 1999. Magnetic Resonance quality control protocols provided by the manufacturer or by Imaging: Physical Principles and Sequence Design, Wiley-Liss, independent sources (e.g. EUROSPIN). MR image quality is New York; McRobbie, D. W., E. A. Moore, M. J. Graves and characterised by several factors, including spatial resolution, M. R. Prince. 2002. MRI from Picture to Proton, Cambridge image contrast, signal to noise ratio and the presence of image University Press, Cambridge, UK; NessAiver, M. 1996. All artefacts. You Really Need to Know about MRI Physics, Simply Physics, Spatial Resolution: Spatial resolution defines the small- Baltimore, MD. est detectable detail of an image. The smaller the voxels of an image, the higher is the resulting spatial resolution. The volume Image quality indicators of a voxel depends on the image matrix size (e.g. 256 × 256), the (Diagnostic Radiology) As the overall aim in medical imag- field of view (FOV) and the slice thickness. Because only a finite ing is the generation of an image that is adequate to answer part of k-space is covered during the imaging process, and various the clinical question, the evaluation of image quality is a cru- other physical limitations, the actual spatial resolution is usually cial element of quality performance. Image quality is assessed poorer than the theoretical pixel limit. both as part of the quality assurance practices, usually utilising I Signal to Noise Ratio: The signal to noise ratio (SNR) is phantoms, but also as part of the evaluation of clinical/patient used to define the relative contributions to an image of the images. detected signal and random noise. Several different definitions Several different indicators can be utilised for the evaluation of SNR are in use, but a common one is the ratio of the average of the image quality, both in planar and in cross-sectional imag- signal intensity within a region of interest (ROI) to the standard ing, aiming to quantify the three main components that affect deviation of the noise within that ROI. SNR can be improved

image quality, that is contrast, unsharpness and noise. TABLE I.1 A Summary of Artefacts in MRI, Their Causes and Appearance Artefact Type Cause Appearance Motion Random or periodic movement of the imaged tissue Periodic: repetitive ghosting Random: image blurring Susceptibility Difference in susceptibility between adjacent tissues or between tissues Signal loss due to dephasing at boundary and air Metal Difference in susceptibility between metal implants/foreign bodies and Signal loss due to dephasing, severe image distortion surrounding tissue Flow Movement of blood through imaged region Changes in signal intensity, ghosting Chemical shift Difference in Larmor frequency between fat and water Spatial offset between fat and water images Gibbs Insufficient sampling of k-space Ringing at sharp contrast steps within image Aliasing Inadequate FOV size, so Nyquist criterion is not satisfied Fold-over of body parts outside FOV back into image Image quality indicators 467 Image quality indicators viewed by the observer. It depends, further to the actual image contrast, on the characteristics of the display and any image processing that may have occurred. Unsharpness: In medical imaging, several components can cause blurring of the final image. Blurring is the diffusion of the signal that smoothens sharp edges and makes them appear more obscured. Quantifying this unsharpness is essential for the evalu- ation of the image quality as it relates to the capability of the sys- tem to properly depict and resolve small and objects. In practice, unsharpness can be quantified: i. Using the limiting resolution, as observed in relevant images of suitable bar patterns, as shown in the image. Influence of contrast unsharpness and noise over real image quality. Contrast: Contrast is defined as the ratio of the signal dif- ference to the average signal. The rationale behind this is that a small difference is negligible if the average signal is large, while the same small difference is readily visible if the average signal is small. In general, in medical imaging, we will want to achieve a high contrast to visualise disease features well. There are two different common approaches to quantify contrast: i. Local contrast (or Weber contrast), is typically used to quantify the visibility of a relatively small object in a relatively uniform background and is defined as: I f - C = f fb Test object for assessment of limiting resolution. fb ii. Using the modulation transfer function (MTF) of the Where ff is the signal of the feature studies and fb is the system, starting from the basics of point spread func- signal of the background. tion (PSF), line spread function (LSF) and edge spread ii. Modulation contrast (or Michelson contrast), used for function (ESF). patterns where both bright and dark features take up similar fractions of the image, is defined as: Noise: Noise is an inevitable component of the medical image and its characteristics need to be quantified. Noise arises as random f - f CM = max min variations in the recorded signal from pixel to pixel. Although fmax + fmin several elements of noise can be defined, the basic source of noise Several different types of contrast can be defined in medical in medical imaging is (or should be) the quantum noise that arises imaging and can be quantified using the two previous equations: from the random generation of the image signal and is related to the number of x-ray quanta. i. Subject contrast, which is the actual contrast of the Noise can be quantified either using simple metrics, such as the feature image, relative to the background, and depends standard deviation of a uniform area or utilising more advanced directly to the physical properties of the two materials, metrics, such as the noise power spectra, that can fully analyse such as the attenuation coefficients in the case of planar the noise properties of a system and can investigate sources of x-ray imaging. detector noise. ii. Image contrast, which depends not only on the subject Other Image Quality Indicators: Combining the indicators contrast but also on the detector and its ability to depict that are directly related to the contrast, unsharpness and noise, the contrast differences. more complicated indicators can be derived that can better char- iii. Display contrast, which goes beyond the image contrast acterise the quality of the image and can be more related to the and actually describes how the contrast is displayed and actual technique or clinical problem. Image reconstruction 468 Image selected in vivo spectroscopy (ISIS) Some typical indicators used in medical imaging are signal to acquisition time, administered activity, etc. One of the earli- noise ratio (SNR), contrast to noise ratio (CNR), detective quan- est methods suggested was filtered back projection. The image tum efficiency (DQE), detectability index (d’), etc. data are transformed into Fourier space and filtered, then trans- Computed Radiography, Digital detectors formed back to the spatial dimension where it is reconstructed Further Reading: International Atomic Energy Agency. 2014. according to the back projection theorem. This reconstruction Diagnostic Radiology Physics: A Handbook for Teachers and method produces an adequate result with low computational Students, Vienna, Austria. power required. Acquired data do not only consist of ‘true’ events but are also Image reconstruction ‘contaminated’ with false events due to a number of reasons, e.g. (Magnetic Resonance) This term relates to the process of convert- scattered radiation and noise from electric circuitry. These effects ing raw data that have been acquired using an MRI scanner into can be suppressed in the reconstruction if the effect can be mod- an image. elled. One example of a reconstruction method where it is possible In conventional Fourier imaging, raw data take the form of to remove false events is called Iterative reconstruction. a two-dimensional array of complex data in k-space. Each item An iterative reconstruction is started with a ‘guess’ or an of data in the array represents the vector sum of magnetisation estimation of the acquired image. Projections from this esti- throughout the excited slice following a specific phase encod- mated image are derived and compared to the actual mea- ing episode and at a specific point during the application of the sured projections. Using information from this comparison frequency encoding gradient. Each data point therefore contains the estimated image is updated and the procedure is repeated. information about the entire image, but the phase informa- Iterative reconstruction was considered to be too computation- tion encoding position in these data points cannot be extracted ally heavy but recent advances in computer technology has led directly. to a much greater use of this method. Two different types of The signal element at a given location in k-space is given by iterative reconstruction are OSEM (ordered subset expectation maximisation) and MLEM (maximum likelihood expectation S (k k ) = òòòr( x y z) -2piéëk ù , xx+k maximisation). yy x , û y , e dx dy dz Related Articles: Iterative reconstruction methods, Filtered back projection, Backprojection reconstruction where ρ(x,y,z) is proton density – essentially the spatial distribu- Further Reading: Cherry, S. R., J. A. Sorenson and M. E. tion of water protons that we wish to recover (in general, weighted Phelps. 2012. Physics in Nuclear Medicine, 4th edn., Elsevier by relaxation times and other parameters, depending on the pulse Saunders, Philadelphia, PA, pp. 253–277. sequence). It can be seen that two-dimensional Fourier transfor- mation of the entire array will yield the desired information. Image registration In some applications, non-cartesian k-space trajectories are (Nuclear Medicine) See Registration used, and these require special treatment in terms of reconstruc- tion. In radial imaging, a series of radial lines through k-space is Image registration acquired by applying two orthogonal gradients simultaneously, (Radiotherapy) This is a multi-modality image processing proce- I the relative amplitude of the two gradients determining the angle dure. In radiotherapy medical images are prepared in the required in k-space of a particular acquisition. The image can be recon- format for treatment planning. This usually involves manual or structed by backprojection, or more commonly by interpolating automatic fusion of CT-MR-PET images. the radially-acquired data onto a Cartesian grid and performing Related Article: Deformable image registration (DIR) two-dimensional Fourier transformation as usual. Similarly, data acquired using a spiral k-space trajectory must Image retrieving undergo Cartesian regridding prior to Fourier transformation. In (General) Image retrieval is the process of recovering a particular partially parallel imaging, the redundancy inherent in data col- image or set of images from a large data archive of images. lected using multiple receiver elements is exploited, either in Within imaging departments, retrieval is usually limited to k-space or in image space, to reduce the required number of phase operating the software of the digital archive of medical images encoding steps (and hence image acquisition time) without sac- stored locally. rificing spatial resolution or field of view (FOV). Image recon- Image data are typically stored within a separate physical struction using these techniques requires additional mathematical electronic data bank (PACS), with each image carrying associated steps either to fill in omitted parts of k-space prior to Fourier details such as type of image, imager settings, and all relevant transformation or to recover useable images from severely aliased patient details. images with restricted FOVs afterwards. Often such image archives are linked to the patient’s elec- Related Articles: k-space, PPI (partial parallel imaging), Raw tronic record, so that when retrieving specific images in clinics, data on the ward, or in the surgery, they can be identified and selected from the patient’s electronic record, and automatically retrieved Image reconstruction and downloaded onto the physician’s monitor. (Nuclear Medicine) This is the process of using acquired 2-D In future, more sophisticated image retrieval systems should projections to build a 3D image volume. The reconstruction pro- be able to assist in research applications. The use of ‘content- cedures are more elaborately described in their separate articles. based image retrieval’ will allow images to be retrieved by their Image reconstruction is used in all tomographic imaging (CT, similarity with other images. MR, PET and SPECT). The goal is to achieve a reconstructed image volume with a high contrast and high signal to noise ratio Image selected in vivo spectroscopy (ISIS) (SNR). These two parameters are not only affected by the choice (Magnetic Resonance) ISIS is one of the most common spatial of reconstruction method, but by a number of factors, such as localisation techniques used for single voxel spectroscopy (SVS). Image sequences 469 Image-guided brachytherapy 180° 180° 180° 90° Related Articles: Point resolved spectroscopy (PRESS), STEAM, Magnetic resonance spectroscopy, Single voxel spectroscopy RF Further Readings: Keevil, S. F. 2006. Spatial localization in nuclear magnetic resonance spectroscopy. Phys. Med. Biol. 51:R579–R636; Ordidge, R. J., A. Connelly and J. A. B. Lohman. G 1986. Image-selected in vivo spectroscopy (ISIS): A new tech- s nique for spatially selective NMR spectroscopy. J. Magn. Reson. 66:283–294. Gy Image sequences (Diagnostic Radiology) Image sequences are used in various Gx x-ray examinations to record quick dynamic changes (most often during x-ray angiographic examinations). One sequence (series) could have approximately two to eight images (exposures) per FID second. This requires powerful x-ray generator capable of pro- ducing short exposures with sufficient power. Often the sequence FIGURE I.11 ISIS pulse sequence. (From Keevil, S.F. 2006. Phys. Med. of exposures is produced by grid-controlled x-ray tube. The image Biol., 51, R579.) sequence can be synchronised with the angiographic injector or with the ECG (in cardio angiography), this way specific phases of the examination (as arterial phase, capillary phase, venous phase) Because it involves acquisition of a free induction decay (FID) can be recorded. One examination can be programmed to have signal shortly after excitation, rather than a delayed echo, it is par- a number of image sequences separated by pauses. The image ticularly suited to nuclear species with short T2 relaxation times, sequence (known also as serial exposures) can be recorded either such as phosphorus (31P) nuclei. with a spot film camera (in older systems) or with a digital detec- The ISIS technique requires post-acquisition combination of tor (in newer systems as DSA, where the images per second can FID signals acquired following each of eight dissimilar pulse be more than 30/s). sequences. One of these sequences is shown in Figure I.11. The Related Articles: Serial exposures, Digital subtraction angi- effect of this sequence is to invert magnetisation throughout three ography (DSA) mutually orthogonal slices, so this magnetisation lies

along the negative z-axis. Where two slices intersect, magnetisation expe- Image smoothing riences two inversion pulses and hence is returned to the posi- (Nuclear Medicine) Image smoothing is a special case of filtering tive z-axis. In the volume formed by the intersection of all three performed to reduce the impact of noise. Since noise is a high slices, magnetisation experiences three inversion pulses and the frequency phenomenon most of the image smoothing filters are net effect is to place it along the negative z-axis. Signal acquired low-pass filters that enhance low frequencies relative to higher at the end of the sequence using the 90° pulse contains positive frequencies. The prominent downside when using smoothing fil- I and negative components reflecting this pattern of spin prepara- ters is the loss of spatial resolution. tion. The other seven sequences are obtained by omitting slice selection along each of the three axes in turn (three sequences), along each possible combination of two axes (three sequences) Image storage and along all three axes (one sequence). After each sequence, (Diagnostic Radiology) See Picture archiving and communica- the spatial distribution of positive and negative z-magnetisation tion systems (PACS) throughout the sample, and hence the composition of the FID signal, varies. Appropriate addition and subtraction of all eight Image uniformity signals should ideally lead to cancellation of signal outside the (Magnetic Resonance) Image uniformity is a measure of the volume of interest (VOI) defined by the three intersecting slices, extent to which an MR image faithfully reproduces a perfectly leaving only the VOI signal for Fourier transformation to yield the uniform imaged object. It reflects the performance of the imaging desired spectrum. system as a whole: hardware, imaging pulse sequence and recon- In practice, the process of adding and subtracting eight large struction software, and can also be affected by features of the signals to produce a relatively small localised signal is subject to object such as magnetic susceptibility differences and dielectric a variety of sources of error. Variation in signal intensity due to properties. See Non-uniformity movement or instrumental drift can result in contamination of the Related Article: Non-uniformity spectrum with signal from outside the VOI, and there can also be dynamic range and digitisation problems. If the read-out pulse is Image-guided brachytherapy not precisely 90° over the whole sample and the sequence repeti- (Radiotherapy, Brachytherapy) In image-guided brachytherapy, tion time is too short for complete T1 recovery, further contamina- IGBT, 3D images of the patient with the applicator(s) inserted tion arises due to a mechanism known as ‘T1-smearing’. Under are used to define tumour extent, target volumes, organs at risk, clinically realistic conditions, T1-smearing can result in up to and the applicator(s) with source and stop (dwell) positions, as the 70% of the acquired spectrum originating from outside the VOI, basis for the treatment planning. although this can be reduced dramatically by judicious choice of In conventional external beam radiotherapy set-up uncertain- VOI and acquisition parameters. Several variants of ISIS have ties, etc. are compensated for by margins added to the CTV, the been developed over the years, often with the aim of reducing clinical target volume, resulting in the planning target volume, the contamination. PTV, a purely geometrical concept. Image-guided radiotherapy 470 Image-guided radiotherapy The situation in brachytherapy is different since the PTV surgeon, oncologist, medical physicist), and channels for the cath- can often be regarded as being the same as the CVT because the eters inserted into the mould. A preplan was also produced before inserted applicators are stable with respect to the CTV. the patient’s treatment to ensure that the geometry and overall The imaging technique chosen for IGBT should in principle treatment plan were correct. be able to show, in a single 3D image data set, the tumour and For the first brachytherapy fraction, a 3D CT image with the the target volume(s), organs at risk as well as the applicators and applicator in place was used to define the target volume, the appli- the source stop (dwell) positions. Imaging techniques must be cator, catheter and corresponding source stop (dwell) positions. A adapted to suit the requirements of the specific brachytherapy treatment plan was created and accepted, and the patient treated. techniques used and organ of interest, e.g. T2-weighted MR imag- Orthogonal radiographs were also taken and used to verify appli- ing for cervix cancer tumour definition and MR and CT compat- cator position in relation to bony structures for the remaining ible applicators must be used. It should also be noted that accurate three fractions. determination of the dose distribution, faithful applicator recon- Abbreviations: CT = Computed tomography, CTV = Clinical struction and representation of source stop (dwell) positions all target volume, HDR = High dose rate, MR = Magnetic resonance depend on the selection of appropriate imaging techniques. and PTV = Planning target volume. Generally speaking, for HDR brachytherapy with the applica- Related Articles: Volumetric prescribing – brachytherapy, tors firmly held in place during the whole procedure, the move- Interstitial brachytherapy, Interactive implant technique ments of applicators relative to target and organs at risk should be minimal, i.e. the dose delivered should correspond to the dose Image-guided radiotherapy planned. (Radiotherapy) Image-guided radiotherapy (IGRT) is a term used Figure I.12 shows a simple example of an image-guided to describe several approaches to using imaging in radiation ther- brachytherapy using CT. This is an intracavitary treatment of a apy. The first of these is the use of functional imaging, particu- residual maxillary cancer using a mould technique with three larly SPECT/PET and MRI, to aid target volume definition. The catheters. The patient had been treated previously with pre-oper- second is the use of soft tissue image to aid verification at time of ative external beam radiotherapy and surgery, and the residual treatment for external beam radiotherapy. The third is the use of tumour is being treated with brachytherapy. The mould had been anatomical imaging to assist in the planning and source implanta- fabricated specifically for this patient and tested (it was inserted tion for treatments such as brachytherapy. in place in the maxillary cavity through the mouth, as the floor Image-Guided Planning: Image-guided planning involves of the mouth had been removed at surgery), optimal catheter the use of functional positron emission tomography (PET), single positioning was discussed by the treatment team (head-and-neck photon emission computed tomography (SPECT) or magnetic I FIGURE I.12 Image-guided brachytherapy: Images of a patient with a maxillary cancer, previously treated with pre-operative external beam radio- therapy and surgery, show a brachytherapy insertion and dose distribution to the residual tumour. A three channel mould applicator has been used with CT-based treatment planning (BrachyVision, Varian). Imaging 471 Immobilisation resonance imaging (MRI) to help define the target volume for Imaging radiotherapy planning. Often the functional tracers used are (General) Imaging is the process of producing visual representa- markers for metabolism (as in 18F-FDG PET imaging). More tions of physical objects. Medical imaging is the process of pro- recently markers for hypoxia (e.g. 60Cu-ATSM PET) and cell pro- ducing images of internal body anatomy, conditions and functions liferation (e.g. FLT PET) have been developed and have started using a variety of medical imaging modalities. to be used. Often the data from these functional modalities are combined with anatomical imaging using CT to maximise the Imaging biomarkers information used to define the target volume. Radiotherapy plan- (General) A biomarker is a biological characteristic that is objec- ning is done on CT data and all other imaging modalities have to tively measured and evaluated as an indicator of normal or patho- be registered to the CT so as to give clinicians information to help logical processes. An imaging biomarker is a characteristic that them define the target volume. is detectable on an image that is usually obtained from PET, Image-Guided Treatment Verification of External Beam MRI and CT scans and comprises measurements of structural Treatments: Several approaches exist to enable image-guided or metabolic features of the body. Imaging biomarker includes radiotherapy for verification. These broadly fall into two catego- the standardisation and optimisation of imaging acquisition pro- ries: the use of internal fiducial markers as a surrogate for the tocols, data analyses, display methods and reporting structures. tumour (imaged with kV or MV x-rays in portal imaging); and the Biomarkers can be anatomical or functional, qualitative or quan- imaging of soft tissue directly. The latter is often achieved using titative. Several imaging biomarkers can contribute to a single or conebeam CT, or CT on rails, or using ultrasound – all in the several response criteria. Reliability of biomarkers is based on treatment room. Another approach to the fiducial marker method accuracy, precision, normal patient variability, correlation with is to use implanted electromagnetic transponders whose positions diseases and usability. The process of acquiring proper images is are located using a receiver. Figure I.13 shows an example of a complex. Standardisation and progress towards better reproduc- conebeam CT imaging system for IGRT verification. ibility will require a coordinated effort by imaging device and Image Guidance in Brachytherapy: Image guidance may be software manufacturers, regulatory bodies, healthcare providers, used in two ways in brachytherapy. Firstly CT scanning may be academic institutions, groups using imaging in clinical trials and used to plan the source positioning and resultant dose distribution. professional societies. Secondly ultrasound or x-ray projection imaging guidance may be used to aid positioning of the sources or applicators under surgery. Imaging system Abbreviations: CT = Computed tomography, Cu-ATSM (Diagnostic Radiology) Imaging system is a general term used to = Cu-di acety lbis( N(4)m ethyl thios emica rbazo ne), FDG = describe all components of imaging equipment – from the cre- Fluorodeoxyglucose, FLT = 3-deoxy-3-fluorothymidine, F-MISO ation of the invisible ‘latent’ image by physical interactions of = Fluoro-misonidazole, IGRT = Image-guided radiotherapy, various forms of energy with the internal structures within the kV = Kilovoltage, MRI = Magnetic resonance imaging, MV = human body to the displayed image available for viewing, This Megavoltage, PET = Positron emission tomography and SPECT = can include radiation sources, detectors and receptors, scanning Single photon emission computed tomography. procedures, image reconstruction and processing, management Related Articles: PET, SPECT, MRI, Radiotherapy planning, and display. Functional imaging, Treatment verification, Electronic portal imaging, CT, Ultrasound, Brachytherapy, Conebeam CT I Imatron (Diagnostic Radiology) Imatron is a vendor name of electron beam CT (see the eponymous article). Related Article: Electron beam CT Immobilisation (Radiotherapy) Immobilisation (or fixation) of the patient is an essential aspect of radiotherapy, enabling the accurate reproduc- tion of the treatment position. This will reduce the dose to normal tissue, allowing the prescribed dose to be escalated for greater tumour control. A simple example of immobilisation is the use of a rigid couch surface continually throughout imaging, planning and treatment. Routines such as voluntary bladder emptying can be an effec- tive and practical way of minimising internal changes in organ placement. Positioning systems also play an important part, and use skin contour systems, body tattoos, or a reference image using non-invasive infra-red or video based camera systems. Immobilisation devices can either be patient specific devices such as thermoplastic shells, or standard devices that can be tailored to a particular patient by rotation and angling. The immobilisation device must be chosen carefully at the beginning of treatment planning, as the choice must be comfort- FIGURE I.13 Conebeam CT system for IGRT verification attached to a radiotherapy treatment machine. The treatment head is at the top-right, a able and easy to use, as will be used continuously all throughout standard MV portal imaging system is at the bottom-left, the x-ray source imaging, simulator and treatment. Evaluation studies of immobil- for the scanner is at the bottom-right and the conebeam detector is at the isation devices allow determination of volume margins for plan- top-left. (With kind permission of Elekta Synergy®.) ning, and action levels for verification processes such as portal Immobilisation device 472 IMPT (Intensity modulated proton therapy) imaging. Immobilisation is especially important with the rise of implants should be avoided. Implants that involve magnets such IMRT, where the conformal dose distribution creates steeper dose as magnetic sphincters, stoma plugs, dental implants, etc. can be gradients, often close to OARs. Patient set up errors will therefore demagnetised by the MR procedure. They should be removed have a greater effect on clinical outcome. prior to

the examination. Active implants are generally consid- One of the most sophisticated immobilisation devices com- ered an absolute contraindication for MRI due to the risk of severe monly used are stereotactic frames which give <1 mm accuracy patient damage or death. The screening procedure for patients is and are used for radical brain radiotherapy. one of the most critical components of a program to permit the Related Articles: Portal imaging, Set-up error, Stereotactic safety of all those preparing to undergo MR procedures or to enter frame the MR environment. It should be noted that having undergone a previous MR procedure without incident does not guarantee a safe Immobilisation device subsequent MR examination. Various factors (e.g. static magnetic (Radiotherapy) See Immobilisation field strength of the MR system, orientation of the patient, orien- tation of a metallic implant or object) can substantially change IMPCB (International Medical Physics Certification Board) the scenario and therefore, a comprehensive screening procedure (General) The International Medical Physics Certification Board must be conducted any time a patient prepares to undergo an MR (IMPCB) is an organisation supported by the International procedure. Organization for Medical Physics (IOMP). IMPCB was formed Related Article: Metallic implant in 2010 to address the global need for certified medical physicists. The objectives and purposes of IMPCB are: Implant dose distribution (Radiotherapy, Brachytherapy) The dose distribution for an inter- • To support the practice of medical physics through a stitial implant or intracavitary insertion should be based on true certification programme in accordance with IOMP position of applicators/sources in relation to the volumes of inter- guidelines; est, i.e. tumour, target and organs at risk. • To establish the infrastructure, requirements and Film Based Implant Dose Distributions: When dosim- assessment procedures for the accreditation of medical etry systems are used for predictive treatment planning, the physics certification programmes in accordance with actual applicator and source positions must be determined and the requirements of IOMP guidelines; approved before the treatment is started. Radiographs are gen- • To establish the infrastructure, requirements and erally used to determine the position of sources and the mark- examination procedures for the certification of medi- ers used to identify organs at risk, if any, in 3D. At least two cal physicists in accordance with the requirements of projection radiographs with known geometry are necessary to IOMP guidelines; define a point in three dimensions and orthogonal radiographs • To provide guidance and support to medical physics are most commonly used for the geometric reconstructions. It organisations for the establishment of national medical is important to note, that soft tissues cannot be visualised on physics certification boards and to conduct board exam- these films. Starting from a reconstruction of applicators or inations for medical physicists in countries that have not sources, the implant dose distribution can be calculated in 3D I yet established certification boards; using a treatment planning system with a brachytherapy source • To grant and issue certificates in the field of medical model. All brachytherapy dose distributions are calculated in physics to applicants who have been found qualified by 3D, even if the source model itself is nominally characterised as the Board; one dimensional. • To maintain a registry of holders of such certificates, Image-Guided Implant Dose Distributions: In modern which can be accessed online free of charge by the image-guided brachytherapy a 3D image data set is acquired with IOMP, and to serve the public by preparing and furnish- the applicators in treatment position. This allows simultaneous ing lists of medical physicists who have been certified 3D definition of tumour, target, organs at risk, applicators and by the Board; and source positions. The 3D dose calculations make it possible to • To establish that continuing education and profes- evaluate the implant dose distribution in terms of dose volume sional development are required for certified medical histogram parameters of tumour and target volumes as well as physicists. organs at risk. Although brachytherapy source models still have • Towards the objectives, IMPCB has started to build a number of limitations, image-guided brachytherapy represents models to develop national certification programmes, a great advance. has established requirements for successful completion Abbreviation: 3D = Three dimensional. of the certification process and is working on collabora- Related Articles: Treatment planning systems – brachyther- tions with IOMP and IAEA. apy, Source models, Orthogonal films, Dose volume histograms – brachytherapy Related Articles: IOMP Further Readings: Baltas, D., L. Sakelliou and N. Zamboglou. Hyperlinks: www .impcbdb .org 2007. The Physics of Modern Brachytherapy for Oncology, Taylor & Francis Group, Boca Raton, FL; Venselaar, J. and J. Pérez- Implant Calatayud, eds. 2004. A Practical Guide to Quality Control (Magnetic Resonance) Metallic passive implants may cause seri- of Brachytherapy Equipment, ESTRO Booklet No 8, ESTRO, ous effects, which include torque, heating on the implants and Brussels, Belgium. artefacts on MR images. Before imaging patients with MR any surgical procedure that the patient underwent prior to the MR examination must be assessed and the presence and character- IMPT (Intensity modulated proton therapy) istics of the implants have to be determined. Ferromagnetic (Radiotherapy) See Intensity modulated proton therapy (IMPT) Impulse response function 473 In silico Impulse response function In silico (Nuclear Medicine) The impulse response function (IRF) (Radiotherapy) In silico modelling refers to the use of computers describes imaging systems degradation because of its inherent to simulate scientific processes. limitations. Another common word is the point spread function In silico methods can model highly complex biological sys- (PSF) or point response function (PRF). The idea behind deter- tems (e.g. cellular proliferation, communication, interactions with mining the degradation is to image a source that is much smaller reagents) that cannot be easily described by sets of equations. than the spatial resolution of the system. For SPECT systems Instead, stochastic methods are often used, drawing upon random and scintillation cameras, one often uses a sealed source such as number generators and probability distributions. Co-57. The full width at half maximum (FWHM) is determined For example, tumour growth algorithms use input parameters from the image of the source and will provide information about and rules to model cell progression and propagation on the micro- the system spatial resolution. The modulation transfer function is scopic scale. In other words, they combine biological input data calculated from the Fourier transform of the IRF and may some- with computer-based models of biological systems in order to times provide a better understanding of the degradation in image conduct investigations of hypotheses entirely in a computer (in quality. silico) environment (Figure I.14). In silico modelling has been applied in a number of areas of Impulse response function radiation oncology and radiation biology, for example to: explain (Diagnostic Radiology) The impulse response function (IRF) clinical observations (e.g. to compare trials); predict clinical out- describes an imaging system or, more often, an image process- comes under conditions not previously measured (e.g. alterna- ing technique under the assumptions of both linearity and shift tive fractionation schedules); optimise radiotherapy treatments invariance. The IRF is used to develop a general input/output (e.g. combined chemo-radiation); identify and evaluate radiation relationship of the system. Knowing the IRF, the system out- risks (e.g. long-duration space flight, deterministic radiation side put for any input is the convolution integral with the impulse effects). response. The advantages of modelling in cancer research are as follows: When applying this concept to an imaging system, a most (i) modelling can significantly reduce the time required to answer common definition is the point spread function (PSF) or point a question and the costs associated with research; it can also response function (PRF), which basically shows the limitations of reduce risks, e.g. in a clinical trial. (ii) Modelling permits easy the imaging system, mainly in terms of spatial resolution. investigation of ‘what if?’ scenarios; that is, models have predic- This concept is more often applied to an image processing tive power. (iii) Modelling allows for the quantitative assessment system, such as e.g. a digital linear and shift invariant filter. The of qualitative processes. (iv) Due to the low costs involved, mod- IRF fully characterises the digital filter. This means that the out- elling allows researchers with modest infrastructure to contrib- put of the filter can be mathematically computed as the discrete ute valuable ideas to the field. (v) Modelling enables researchers convolution sum of the input image and the IRF or, equivalently, to quantify and interpret experimental results, including data by exploiting the convolution theorem which means using Fourier obtained from clinical trials. Transformed signals. This approach does not hold for more The disadvantages are as follows: (i) It is very difficult, if refined image processing techniques such as non-linear or adap- not impossible, to model the entirety of a biological process. (ii) tive filters. There is always a compromise in a modelling process as only a I Related Articles: Point spread function (PSF), Convolution certain number of known biological processes/parameters can integral, Linear and shift invariant systems be accounted for within the model. This known compromise is in addition to an unknown compromise: unknown processes may occur in the modelled biological system. (iii) The results of In-111-labelled ibritumomab tiuxetan (Zevalin™®) an in silico model should be considered clinically-relevant only (Nuclear Medicine) See Y-90-labelled itbritumomab tiuxetan (Zevalin®) In air calibration factor (Radiation Protection) The ionisation chamber is recommended by international protocols for determination of the absorbed dose in photon and electron beams. In air calibration factor relates the dose in the chamber gas to the measured collected charge. The determination of this calibration factor for the parallel plate ioni- sation chamber is obtained from a comparison of the absorbed dose to water value measured with a cylindrical reference cham- ber, as recommended by international protocols IAEA TRS 381, IAEA TRS 398. Abbreviations: IAEA = International Atomic Energy Agency and TRS = Technical reports series. Related Articles: Dose, Radiation dosimetry Further Readings: Absorbed Dose Determination in External Beam Radiotherapy–2000 An International Code of Practice for Dosimetry Based on Standards of Absorbed Dose to Water. Technical Reports Series No. 398, IAEA; IAEA. Absorbed Dose Determination in Photon and Electron Beam – An International FIGURE I.14 Schematic diagram showing computational (in silico) Code of Practice for Dosimetry. Technical Reports Series No. 381. modelling drawing on experimental data and theoretical knowledge. In vitro 474 In vivo body composition after the credibility of the model has been validated by in vivo The body compartments are as follows: total body protein results. (TBP), total body nitrogen (TBN), total body potassium (TBK), Further Readings: Bezak E, Marcu L, and Penfold P, total body water (TBW), total body fat (TBF), total body calcium Computational and Mathematical Modeling of Tumor Kinetics (TBCa), total body carbon (TBC) and fat free mass (FFM). FFM and Response to Radiation and Chemotherapy, Computational includes all body tissue excluding fat: i.e. skeletal muscle, visceral and Mathematical Methods in Medicine, vol. 2012, Article ID organs, bone and skin, and body water, as well as hair, blood, and 702675, 2 pages, 2012. https://doi .org /10 .1155 /2012 /702675; lymph. Trisilowati and Mallet DG, In Silico Experimental Modeling of Compartments are related as follows: Cancer Treatment, ISRN Oncology, vol. 2012, Article ID 828701, 8 pages, 2012. https://doi .org /10 .5402 /2012 /828701. TBP = 6.25* TBN In vitro FFM is calculated from TBK on the basis that the potassium con- (General) In vitro (from Latin: within the glass) refers to an tent of FFM is 2.26 g/kg in females and 2.52 g/kg in males, as experimental technique performed ‘in a test tube/culture dish’ potassium is not found in adipose tissue. using controlled environment conditions and outside of a living However, changes in TBK can reflect chemical changes in the organism. body and blood. For this reason TBN is regarded as a superior An example would be the investigation of the response of measure of protein status in diseased subjects. cancer cells to radiation, when these cells are grown in flasks or The mass of a subject is the sum of the defined compartments Petri dishes, subjected to irradiation and then cell survival, DNA in the two and four body models: double-strand breaks or other end-points are observed. In vitro experiments represent a very simplified environment. M = FFM + TBF They are, however, often required as first proof-of-principle tests prior to conducting experiments in animal models and/or human M = TBP

+ TBF + TBW + TBCa subjects. While, on one hand, in vitro studies allow the study of phe- Methods: In vivo methods are used to obtain values of the nomena in isolation without contributing and confounding fac- body compartments. These might be done before and after tors of the whole organism, results of such experiments cannot be therapy, to determine the effect of therapy. The measurement directly generalised to living organisms. of normal values is required in order to compare patient values, The clonogenic assay technique is an example of an in vitro according to defined indices. For example, the nitrogen index NI experiment. = TBNp/TBNn, where p designates the patient and n the normal values for healthy subjects. In vivo The importance of such indices relates to the impact on dis- (General) In vivo (from Latin: within the living) means an experi- ease prognosis. mental investigation using a whole, living organism as opposed to In vivo interrogation techniques are used. These can give a cell cultures or partial or dead organism. measure of the whole compartment, or in some cases the spatial I Animal and human studies (including clinical trials) are distribution of the compartment. Analysis of blood and urine after examples of in vivo investigations and research. ingestion of isotopically labelled molecules can also be used: In vivo testing must be conducted to observe the overall effects of an experiment/drug/radiation treatment on a living organ- 1. The major clinical application of in vivo body composi- ism, as it will elicit all the complex contributing and causational tion analysis rests with the dual energy x-ray (DEXA) effects and responses of a biological system rather than those of, technique for the determination of bone density, of for example, isolated cells. particular relevance to the management of osteoporo- In vivo testing and verification must be conducted prior to sis, especially in the aging population. There is a very any new treatment being approved for clinical use. This may low and insignificant radiation dose associated with this include investigation of the maximum tolerated doses as well as, measurement (∼10 μSv). for example, new drug efficacy and side effects. Relevant ethics 2. Computed Tomography (CT) can be used to measure approvals must always be obtained prior to any in vivo studies. bone density but invokes much higher and significant Another term used in biological experiments is ex vivo (from radiation doses. Latin: out of the living), referring to procedures or studies involv- 3. High energy neutron inelastic scattering in calcium can ing living cells or tissue samples taken from a living organism. give the TBCa by measurement of the characteristic These are then cultured in a laboratory apparatus and conditions inelastic gamma ray, but at a high radiation dose. for a short period of time. Experiments lasting longer are then 4. Nitrogen is measured in a body protein monitor typically considered to be in vitro. (BPM) using a Cf252 or PuBe neutron source. The patient is moved over a collimated neutron beam. The In vivo body composition fast neutrons emitted by these radioactive sources are (Radiotherapy) moderated in tissue and captured by hydrogen and Background: The body comprises distinct and measurable nitrogen nuclei in the patient. The 11.4 MeV ground body compartments. The status and rate of change of these com- state gamma ray from nitrogen capture is weak but partments reflects the health of a person and the response of treat- can be measured by NaI detectors as I has higher ment for a specific disease. Whereas measurement of body weight energy than all background radiations. The hydrogen (M) is a basic and useful parameter, it can be a misleading mea- gamma ray is 2.2 MeV but the yield is intense and eas- sure of response to treatment, which may increase oedema and fat ily measured. The ratio of N/H is independent of body while depleting protein. habitus. In vivo dosimetry 475 Incident analysis framework 5. Potassium-40 has a half-life of 1.3 × 109 years, and one treatment session. Portal dose measurements are extremely is present with 0.0117% abundance in all potassium. useful in detecting differences between actual patient data as The 1.46 MeV gamma ray is readily detected by an encountered during treatment and those applied during treatment array of NaI detectors. There is no additional radiation planning. EPIDs are likely to become very useful for dosimetric dose, apart from the naturally occurring potassium. A quality assurance of intensity-modulated beams. low background, shielded room is required for accu- For in vivo dosimetry, the most commonly used detectors are rate measurements with old steel (forged prior to the silicon diodes, MOSFET dosimeters and TLDs. A-bomb testing program). Abbreviations: EPID = Electronic portal imaging device, MOSFET = Metal oxide semiconductor field-effect transistor, Abbreviations: BMI = Body mass index, BPM = Body protein SSD = Source surface distance, TBI = Total body irradiation, monitor, DXA (previously DEXA) = Dual energy x-ray absorp- TLD = Thermoluminescent dosimeter and TPS = Treatment tiometry, FFM = Fat free mass, LBM = Lean body mass, NI = Planning System. Nitrogen index, TBC = Total body carbon, TBCa = Total body Related Articles: Entrance dose, Exit dose, Dose verification, calcium, TBF = Total body fat, TBK = Total body potassium TBN Diode detectors, TLD = Total body nitrogen, TBP = Total body protein, and TBW = Further Readings: Podgorsak, E. B. 2005. Radiation Total body water. Oncology Physics: A Handbook for Teachers and Students, IAEA, Related Articles: Total body protein, Total body nitrogen, Vienna, Austria; Vam Dam, J. and G. Marinello. 2006. Methods Total body potassium, Total body water, Total body fat for In Vivo Dosimetry in External Radiotherapy, ESTRO Booklet No. 1, ESTRO, Brussels, Belgium. In vivo dosimetry (Radiotherapy) In vivo dosimetry is the verification method for In vivo range verification checking dose delivery directly during the treatment. It is per- (Radiotherapy) In proton and hadron therapy, uncertainties formed to detect errors in individual patients, to detect errors in particle range arise from (i) inaccuracies and degeneracy in in core procedures, to evaluate the quality of specific treatment the stoichiometric calibration of CT Hounsfield units to proton techniques or to evaluate the dose in situations in which the dose stopping powers, (ii) image artefacts (e.g. due to metal implants) calculation is inaccurate or not possible (e.g. non-standard SSD and (iii) anatomical changes during treatment. These uncertain- or using bolus). ties must be considered during treatment planning, ultimately In vivo dose measurements can be divided into entrance dose limiting the precision of the therapy. In vivo range verification measurements, exit dose measurements and intracavitary dose is a means of reducing/assessing particle range uncertainties just measurements. before/after treatment for a specific patient on a specific treatment Entrance dose measurements serve to check the output and day. Various methods for in vivo range verification have been pro- performance of the treatment apparatus as well as the accuracy of posed, including: (i) the direct placement of diodes within body patient set-up. If entrance dose measurements alone are applied, cavities, (ii) prompt gamma, (iii) PET, but using positron emitters the entrance dose has to be converted to the corresponding target directly produced within the body by the treatment beam and (iv) dose using patient and treatment set-up information. radiographic (2D) or tomographic (3D) imaging using ion beams. Exit dose measurements serve, in addition, to check the dose Related Articles: Stoichiometric calibration, Proton CT I calculation algorithm and to determine the influence of shape, (pCT), Prompt gamma, Positron emission tomography (PET) size and density variations of the body of the patient on the dose Inch calculation procedure. In the real patient there is in most cases a (General) The inch is a small Imperial and US unit of length, considerable loss of backscatter, while the TPS calculations are being equivalent to exactly 25.4 mm, and one twelfth of the larger valid for semi-infinite patients implying complete backscatter at unit of length, the ‘foot’. the exit surface. A correction is then necessary. The abbreviations ‘ft’ and ‘in’ are commonly used, as in 3 ft 6 A combination of entrance and exit dose measurements is a in., and also in shortened form as in 3′6″. more accurate method of obtaining the target dose. Various meth- ods are available to obtain the midline dose from entrance and exit dose values. These methods give generally good results for Incidence angle homogeneous situations but in the presence of inhomogeneities (Radiotherapy) See Oblique incidence considerable deviations can occur. Related Articles: Oblique incidence, Electron oblique inci- In vivo dose measurements not only serve to check the dose dence, Obliquity, Obliquity effect delivery to the target volume but are also applied to assess the dose to organs at risk (e.g. the eye lens, gonads and lungs during TBI). Incident Air Kerma Portal imaging systems can also be used for in vivo dosimetry. (Radiation Protection) The air kerma from an x-ray beam mea- Portal images can be transformed to ‘dose images’, which can then sured on the central axis at the patient or phantom surface, exclud- be correlated with exit dose values. Various groups are currently ing backscatter is illustrated by Figure I.15. studying the usefulness of films or EPIDs for in vivo dosimetry. Related Articles: Kerma, Air Kerma Forward approach and backward approach can be used. The rela- Further Readings: Dance, D. R., S. Christofides, A. D. tionship between the exit dose and the transmission dose at the Maidment, I. D. McLean and K. H. Ng. 2014. Diagnostic position of the portal imaging detector is not simple and depends Radiology Physics, International Atomic Energy Agency. on many factors, such as the skin to detector distance, field size, patient thickness and photon beam energy. Incident analysis framework Since a relatively large number of images can be made during (General) The incident analysis framework is the guideline that one treatment fraction, EPIDs can be used to measure the influ- allows researchers to go deeper, in a sequential and organised ence of organ and patient motion on the dose distribution during way, into every aspect concerning adverse events that have Incident dose 476 Indium-111 [111In] associated with hospital light sources with reference to the Control of Artificial Optical Radiation at Work Regulations 2010. J. Radiol. Prot. 30(3):469; Kitsinelis, S. and S. Kitsinelis. 2015. Light Sources: Basics of Lighting Technologies and Applications, CRC Press; Li, T. H. and G. Y. Luo. 2015. Design of light source of agricultural UVALED pest control lamp in food production. Adv. J. Food Sci. Technol. 9(1):36–39. Incoherent scattering (Radiation Protection) Interaction between an incident photon and a loosely bound atomic electron in which only some of the energy of the photon is transferred to the electron. The photon is scattered into a new direction, and the electron, referred to as Compton or recoil electron, is ejected from the atom with kinetic energy equal to the loss of energy of the photon. FIGURE I.15 Diagrammatic representation of incident air kerma. Also known as inelastic scattering, Compton interaction, or Compton scattering. Related Article: Compton effect occurred or that can occur in healthcare facilities related to the use of biomedical equipment or practising inadequate proce- Indirect detection dures. The incident analysis framework is divided into several (Radiation Protection) Photons and neutrons are indirectly ionis- levels and steps (for example, data collection by a suitable num- ing radiation, that is they react with the medium (the matter the ber of analysts in a control room, organisation and data process- radiation is passing through) producing charged particles that are ing by chronological transcription of events with identification ionising. of the various operators and participants, data analysis, detec- For photons, the photon(s) beam passing through matter can tion of results and conclusions) in order to identify the vari- interact with orbital electrons (photoelectric effect), producing ability of performance and the safety of medical devices and photoelectrons, or scatter, losing some of its energy to an orbital operating methods implemented in hospital by technicians and electron which is ejected from the atom (Compton effect). If the clinicians. The framework of the analysis changes in case of photon energy exceeds 1.022 MeV (i.e. the sum of rest mass of a priori identification of potential adverse events (prospective electron and positron) the photon may interact with the atomic study) and the investigation of those that occurred (backward

nuclei and an electron–positron pair created. analysis). This incident analysis allows the identification of criti- Neutrons are detected using nuclear reactions such as (n, α), cal issues within a company or a healthcare facility and the pro- (n, p), (n, fission) or neutron activation, in which case a radioac- cesses enacted in order to make improvements and avoid the tive isotope is produced. further onset of adverse events. Also, the indirect detection of ionising radiation is considered I as based on the changes induced by radiation in the detector that Incident dose are not observable immediately (directly). The intensity of radia- (General) This is an ambiguous term used to describe the tion measured for estimating, e.g. radioactivity of the sample or absorbed dose (usually in air) or air kerma or exposure incident dose, may be read after the exposure and not during exposure as it on an absorber. See related terms. is possible with the use of ion chamber, scintillation or semicon- Related Articles: Absorbed dose, Air kerma, Exposure ductor counters. The monitoring of the radiation exposures for safety purposes Incident energy fluence of the staff is often realised with use of indirect (passive) detec- (Radiation Protection) This is not a recognised radiation unit. It is tors, such as film badges where the optical density of the film used to describe the energy fluence incident on an absorber. changes when irradiated. After developing the optical density Related Article: Energy fluence of the film it is measured, which with correct calibration, can be related to the radiation dose. The thermoluminescent detector Incoherent light source (TLD) stores the absorbed energy of ionising radiation and when (Non-Ionising Radiation) An incoherent light source, or non-laser heated visible optical radiation is emitted, intensity of which is source, is one where the irradiation is generated by spontaneous proportional to the radiation dose and can be measured, for exam- emissions made of photons in different energy states and phases. ple using a photomultiplier. Non-laser optical sources generate broad-band spontane- The alanine radiation detector (aminopropionic acid) is a ous emissions from electrically excited materials, which can be dosimeter in which ionising radiation produces free radicals and metals (incandescent sources), gases (gas discharge sources) or concentration of the free radical is measured using the EPR (elec- semiconductor junctions (light-emitting-diodes). These days, tron paramagnetic resonance) technique. The reading does not gas and LED sources are the most common and are the most destroy the record, as in case of TLD. energy-efficient. Neutron detection is based on proton recoil or alpha particles Healthcare settings have plenty of incoherent light sources recorded, for example by track-etch detectors which are special foils ranging from general and examination lighting to therapeutic light saturated with 10B and the reaction 10B(n, α)7Li produces α-particles sources, such as neonatal blue light therapy and UV phototherapy. that mark tracks in the foil which can be seen after a chemical treat- Related Articles: Phototherapy ment. The number of tracks depends on the neutron energy. Further Readings: Coleman, A., F. Fedele, M. Khazova, P. Abbreviations: EPR = Electron paramagnetic resonance and Freeman and R. Sarkany. 2010. A survey of the optical hazards TLD = Thermoluminescent detector (Dosimeter). Indirect digital radiography 477 Indium-111 [111In] Related Articles: Absorbed radiation, Alpha particles, Clinical Applications: Indium-111 has been used in nuclear Compton effect, Dose, Dosimeter, Film badge, Gamma radiation, medicine since the late 1970s, and has become the standard Ionisation chamber, Neutrons, Pair production, Photoelectric radionuclide for labelling of different biomolecules, such as effect, Scintillation detector, Personal dosimetry blood cells (leukocytes and platelets) pentetreotide, and mono- Further Readings: Kane, S. A. 2003. Introduction to Physics clonal antibodies using different linker molecules such as bifunc- in Modern Medicine, Taylor & Francis Group, London, UK, pp. tional chelates. 246, 251; Saha, G. P. 2001. Physics and Radiobiology of Nuclear Today, the most known 111In-labelled radiopharmaceuticals Medicine, 2nd edn., Springer-Verlag, New York, pp. 186–187; are 111In-pentetroide (111In-OctreoScan™, Mallinckrodt Medical) Stabin, M. G. 2008. Radiation Protection and Dosimetry: An for localisation of primary and metastatic neuroendocrine tumour Introduction to Health Physics, Springer Science+Business cells expressing somatostatin receptors and111In-ibritumomab Media, LLC, New York, p. 161. tiuxetan (111In Zevalin™, Schering AG) for imaging of relapsed or refractory follicular non-Hodgkin’s lymphoma, prior to radionu- Indirect digital radiography clide therapy with 90Y-Zevalin. (Diagnostic Radiology) Indirect digital radiography refers to The chemistry of indium resembles that of iron and gallium, digital systems with flat panel, where x-rays are first converted and a close similarity between indic and ferric ions has been estab- to light, which is then detected by photodiodes or CCD (see the lished. Elimination of indium activity from organs and tissues is article Flat panel detector). extended because of trapping by the plasma protein transferrin. Historically x-ray film radiography has been considered a Ionic 111In3+ injected at pH <3.5, 111In-chloride or 111In-citrate, are direct method, while image intensifier systems have been con- rapidly captured by transferrin in the vascular system. Thus, 111In sidered indirect (the signal passes several transformations). Later activity is distributed heterogeneously in the body and may locally the digital systems have passed through similar divisions – direct contribute significantly to the absorbed dose, since even though digital radiography (x-rays transferred to charge by amorphous indium is a foreign element in our body, many different tissues selenium) and indirect digital radiography (using phosphor and show affinity for indium (transferrin receptors), especially rapid photodiode or CCD). proliferating tissues. Related Article: Flat panel detector Indium-111 [111In] 111 49 In (2.80 d) (Nuclear Medicine) Element: indium (group III-A) EC (100%) Isotopes: 51 < N < 84 Atomic number (Z): 49 Neutron number (N): 62 416.70 keV Symbol: 111I EC Production: cyclotron, e.g. 111Cd(p,n)111In ® 111Cd or hν = 171.28 keV (90%) 112 . Cd( 2 111 EC 2 8 p, n) In ® 111 d Cd I 2.8d 245.42 keV Daughter: 111Cd Half-life: 2.8 days hν = 245.42 keV (94%) Decay mode: EC – decay Radiation: gamma, internal conversion electrons, Auger 0 keV electrons, characteristic x-ray photons 111 48Cd stable Gamma energy: 171.28 keV (90%) 245.42 keV (94%) Dose rate from 1 MBq: 0.0842 μSv/h at 1 m; 842 μSv/h at 1 cm Related Articles: Gallium-67, Ytrium-90 ibritumomab tiux- Absorption (HVL): 0.8 mm lead eran (Zevalin) Biological half-life: 70 days Further Readings: Annals of the ICRP. 1987. Radiation Dose Critical organ: red bone marrow, liver, spleen, testes and to Patients from Radiopharmaceuticals, Biokinetic Models and lymph nodes Data, ICRP Publication 53, Vol. 18, Pergamon Press, Oxford, ALImin (50 mSv): 200 MBq UK; Annals of the ICRP. 1998. Radiation Dose to Patients Absorbed dose (chloride): 0.88 mGy/MBq kidneys, 0.60 from Radiopharmaceuticals, Vol. 28, No. 3, Addendum to ICRP mGy/MBq liver, red bone marrow. Publication 53, ICRP Publication 80, Pergamon Press, Oxford, Effective dose: 0.20 mSv/MBq (111InCl3); 0.40 mSv/MBq UK; Chu, S. Y. F., L. P. Ekström and R. B. Firestone. 1999. The (111In-labelled leukocytes, platelets) Lund/LBNL Nuclear Data Search, [http://nucleardata .nuclear .lu .se /toi/ (accessed 31 July 2012)]; Firestone, R. B. 1999. Table of Isotopes, 8th edn., Update with CD-ROM. [http://ie .lbl .gov /toi .html]; Jönsson, B.-A., S.-E. Strand and B. S. Larsson. 1992. A quantitative autoradiographic study of the heterogeneous activity distribution of different indium-111-labeled radiopharmaceuticals in rat tissues. J. Nucl. Med. 33:1825–1833; Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American Pharmacists Association, Washington, DC. Induced radioactivity 478 Inflow effect Induced radioactivity Related Articles: Macro-CT, Dragonfly CT (Radiotherapy) Radioactivity may be induced in the components of a linear accelerator when they are irradiated by an electron or Industrial radiography photon beam. The induced activity depends on the energy, beam (Diagnostic Radiology) Industrial radiography (aka non-destruc- power and type of the irradiated material. The induced radioac- tive testing – NDT) is used widely in industry for inspection of tivity is related to the nuclear reaction that exhibits a threshold various technological elements – e.g. welding of metal tubes, which can be found in the following references. Three types of ship elements, metal bridges and other structures. There are spe- photon-induced reactions produce most of the activity: the giant cific x-ray tubes developed for NDT purposes, as well as specific photonuclear resonance, the quasi-deuteron effect and the high- films and digital detectors. Industrial radiography can also use energy photospallation reaction. The linac components which gamma sources, linear accelerators, particle accelerators, etc. more probably can be suspected of an activation are beam dumps, The requirements for image quality depend on the specific use of targets, collimator and jaws and compensation filters. Also air is NDT, but usually, radiation safety is not of concern. made radioactive by linear accelerators operating above 10 MeV. The photoneutron reaction inducing an activation of nitrogen and Inelastic scattering oxygen is (Radiation Protection) Interaction between an incident photon and a loosely bound atomic electron in which only some of the 14 N (g,n) N13 and O16 (g,n)O15 energy of the photon is transferred to the electron. The photon is scattered into a new direction, and the electron, referred to as Compton or recoil electron, is ejected from the atom with kinetic N13 and O15 decay β+, respectively, with a half-life of 600 and 122 energy equal to the loss of energy of the photon. s, respectively. Also known as inelastic scattering, Compton interaction, or The activity of N13 and O15 per unit volume of air produced in Compton scattering. the treatment room by a 20 cm × 20 cm beam is given by Related Article: Compton effect (D / 16.7)(P)(1- e-(l+Q /V )Tr (e-(l+Q /V )Ts )) Inferior; caudal C = V ( (General) Directional anatomical terms describe the relationship l + Q / V )l of structures relative to other structures or locations in the body. ‘Inferior’ or ‘Candal’ means under or towards the feet (for where example, the foot is inferior to the knee). C is the concentration (Bq/m3) See Anatomical relationships D is the dose rate at maximum build up depth (mGy/s) P is the production rate of N13 or O15 of 16.7 mGy/s (Bq/kg of Inflow effect oxygen or nitrogen) (Magnetic Resonance) In sequences using two or more spatially Q is the room ventilation rate (m3/s) selective RF pulses, such as the SE and IR sequences, the signal V is the volume of the room (m3) I λ is the decay constant for N13 and O15 (s−1 is sensitive to the transport of spins into and out of the excited ) slice. If the flow is coherent over at least a number of voxels Tr is the irradiation time (s) (macroscopic flow) and has components perpendicular to the Ts is the time after end of irradiation (s) imaging slice, the signal will decrease due to the outflow of spins during the time between RF pulses (wash-out effect, see Further Reading: Howerton, R. J., D. Braff, W. J. Cahill and Flow void). On the other hand, if the repetition time is not too N. Chazan. 1964. Thresholds of Photon Induced Reactions, Rep. long, the inflow of ‘fresh’, non-saturated spins during the rep- UCRL-14006, Lawrence Livermore Laboratory, Livermore, CA. etition time, will increase the magnetisation compared with the static, saturated, spins and thus give an increase in signal – an Industrial CT effect called the inflow effect or the wash-in effect (in early (Diagnostic Radiology) Industrial CT is mainly used in research literature also known as ‘paradoxical enhancement’). These departments and quality control labs to perform inspection competing mechanisms make the modulus signal behaviour scans of manufactured parts. Their primary uses are for the versus flow velocity biphasic and possibly difficult to interpret, detection of defects, wall thickness analysis, structural analysis although, with a rough knowledge of the velocities involved, and dimension control. These are scanners that have not been the experimental parameters could be adjusted so that one designed for medical use and hence low dose rates are not an mechanism is dominant. If, however, only one RF pulse is used issue in this technique. Therefore, especially in the case of the per repetition, as in gradient echo (GRE) sequences, the signal inspection of high-density objects, as high x-ray source voltages decrease due to wash-out will not occur and an enhanced signal that may reach up to 15 MeV are used, shielding for both the will result from the inflow effect. Ideally, for plug flow perpen-

source and the detector becomes significant. Due to these shield- dicular to the imaging slice and perfect slice shape, the signal ing requirements and hence increased system weight in industrial increase will be linear with velocity up to the velocity when all CT, unlike medical CT, it is the sample that is rotated within the spins are exchanged in the slice during TR (see the following scanner. figure). Informed Consent 479 Inherent filtration GRE - TR - GRE Related Articles: AORD, ICNIRP, Light Source Imaging slice Further Readings: Czapla-Myers, J. S., K. J. Thome and S. F. Biggar. 2008. Design, calibration, and characterization of a field radiometer using light-emitting diodes as detectors. Appl. Opt. 47(36):6753–6762; ICNIRP website, https :/ /ww w .icn irp .o rg /en /freq uenci es /in frare d /ind ex .ht ml; Ihrke, I., J. Restrepo and L. Mignard-Debise. 2016. Principles of light field imaging: Blood vessel Briefly revisiting 25 years of research. IEEE Signal Process. Mag. 33(5):59–69; Kitsinelis, S. and S. Kitsinelis. 2015. Light Sources: Basics of Lighting Technologies and Applications, CRC Press, Boca Raton, FL. Infrared light hazard (Non-Ionising Radiation) See Thermal Light Hazard Velocity Infrared radiation (Radiation Protection) Infrared radiation is otherwise known as Related Article: Flow void heat radiation and is the part of the electromagnetic spectrum between visible light and microwaves/radiowaves. The IR spec- Informed Consent trum ranges from a wavelength of 0.7 μm at the red end of the (Radiation Protection) Informed Consent is the result of all visible spectrum, to around 1 mm, and is therefore in the non- processes to ensure that persons who may be exposed to ionis- ionising region of the spectrum. ing radiation, whether for work or as a patient, do so knowingly Related Articles: Electromagnetic radiation, Non-ionising and willingly – i.e. they are given appropriate information on radiation the potential benefits and risks of radiation exposure and decide freely and voluntarily to proceed with the exposure. Inherent contrast Informed Consent is not an explicit requirement in the (Magnetic Resonance) Inherent contrast refers to the contrast IAEA Basic Safety Standards nor in the European Basic Safety properties present in an image without the use of an exogenous Standards Directive. However, it is implicit in the definitions contrast agent. of the ethical values of Dignity and Autonomy set out in ICRP Inherent contrast in an MR image may depend on a wide Report 138. range of variables, including the concentration of water protons In its most general form when applied to any medical pro- and their physicochemical environment (T1 and T2), bulk flow cedure, Informed Consent is implicit in the requirements of and perfusion of blood, and the rate of water molecule diffusion. the World Health Organization’s Declarations of Geneva and These variables, in turn, may reflect a variety of physiological Helsinki to respect the dignity and autonomy of the patient/volun- parameters, such as water content, temperature, brain activation teer. It should be further noted that a patient or volunteer having and tumour angiogenesis. I given consent at the beginning of a procedure is then still allowed The extent to which each of these variables influences the to withdraw their consent at any time. appearance of an image is controlled through pulse sequence Related Articles: Basic Safety Standards, IAEA, EURATOM, design, either by simple changes to sequence parameters (e.g. Justification, ICRP shortening the repetition time, TR, to increase T1 weighting) or by adding sequence elements (e.g. Stejskal–Tanner gradients for Infrared light diffusion sensitisation). (Non-Ionising Radiation) The part of the electromagnetic radia- Thus it is possible to alter the inherent contrast and associated tion spectrum just above the visible (780 nm) until about 1mm. information content of an MR image to an essentially limitless Infrared is subdivided into three categories: extent in order to answer different clinical questions. This capa- bility makes MRI far more flexible than most imaging modalities. • Infrared A (IRA): 780 nm–1.4 µm Related Articles: Contrast agent, Unenhanced image • Infrared B (IRB): 1.4 µm–3 µm • Infrared C (IRC): 3 µm–1 mm Inherent filtration (Diagnostic Radiology) The total filtration of an x-ray beam aims Common natural sources of infrared are the sun and fires. to reduce the unnecessary low energy x-ray photons. It is pro- duced by a combination of two filter components – inherent and added. The inherent filtration is due to existing components of the x-ray tube and housing through which the x-ray beam passes (tube glass, oil in tube housing, beam locator mirror, etc.). After remov- ing the added filtration, the inherent filtration can be measured by means of HVL. Related Article: Total filtration Signal Inhomogeneity 480 I norganic phosphate Inhomogeneity (sometimes defined as mixed-models), the internal clinical engi- (General) Inhomogeneity refers to field, image, etc. which is not neering department stipulates contracts for the coverage of cer- uniform (homogeneous). See specific articles related, for exam- tain devices and keeps full internal coverage on all the rest of the ple to field homogeneity in MRI; image uniformity in diagnostic inventory. The extent of outsourced services in this mixed-model radiology or nuclear medicine; dose distribution in radiotherapy, can vary, from preventive maintenance only to high-level correc- etc. tive maintenance (with first level interventions performed by the internal staff) to fully outsourced services on a selected list of equipment. Inhomogeneity correction factor (Radiotherapy) For accurate dose calculations in a heterogeneous volume such as a patient, corrections must be made for the pres- Inorganic phosphate ence of non-water-equivalent tissue, or inhomogeneities. (Magnetic Resonance) Inorganic phosphate features in in vivo One approach to doing this starts with the dose distribution for phosphorus (31P) NMR spectra. The term encompasses four dis- a homogeneous water equivalent density and then applies correc- tinct species – the phosphate ion (PO4 ) (Figure I.16), the hydrogen tion factors to change this distribution to account for the different phosphate ion (HPO4 ), the dihydrogen phosphate ion (H2PO4 ), tissue densities. In this approach, the dose in the heterogeneous and phosphoric acid (H3PO4). case is given by the following equation: These species exist in pH-dependent equilibrium, and under physiological conditions inorganic phosphate occurs as HPO4 and Dhetero = Dwater ´ ICF (I.1) H2PO4. These ions have different chemical shifts, and so the posi- tion of the inorganic phosphate peak in a 31P spectrum depends on the relative proportions of the two species and hence on pH. Thus Equation I.1 shows how to correct a homogeneous dose distribu- the chemical shift difference between inorganic phosphate and tion for the presence of an inhomogeneity using inhomogeneity phosphocreatine is frequently used to determine intracellular pH correction factors (ICF). according to the following formula: where æ d - d Dhetero is the dose distribution in heterogeneous tissue pH = pK + HA ö A ç ÷ è dA - d Dwater is the dose distribution in a water volume ø ICF is the inhomogeneity correction factor Here δ is the chemical shift of the Pi peak relative to PCr, δA is the chemical shift of hydrogen phosphate (approximately 5.70 ppm), There are a number of methods for calculating ICF including δHA that of dihydrogen phosphate (approximately 3.23 ppm), and approaches such as effective attenuation coefficient, power law pKA = 6.77 (Figure I.17). (Batho), ratio of TAR and equivalent TAR. These are summarised in a number of textbooks and reports including APPM (2004). Abbreviations: ICF = Inhomogeneity correction factor and TAR = Tissue air ratio. O– I Related Article: Heterogeneity O– P O– Further Reading: AAPM. 2004. Tissue inhomogeneity cor- rections for megavoltage photon beams, AAPM Report number O 85, Medical Physics Publishing, Madison, WI. FIGURE I.16 Molecular structure of the phosphate ion. In-house service (General) The management of medical equipment in a health- care institution can be organised through different models and it is important that the people in charge are clearly identified and the corresponding roles and responsibilities are well defined and Inorganic communicated at each level of the organisation. phosphate One of the possible organisation models is the so-called in-house service, where ‘the servicing of medical equipment is performed by the facility’s own staff’ (as defined in the Universal Medical Technology Service Nomenclature [UMTSN] by ECRI Institute). Strictly speaking, this model requires that all technical staff are employed by the healthcare organisation, including clinical engineers, medical physicists, biomedical technicians, etc. This model has both advantages (full control on the internal staff, good leverage on the OEMs for the provision of staff training and spare parts availability, knowledge stays inside the healthcare facil- ity) and disadvantages (low flexibility, complex management of human resources, high start-up costs, difficult training on each different equipment model, yearly costs are not fully predictable). 25.000 20.000 15.000 10.000 5.0000 0.00000 –5.0000 –10.000 –15.000 –20.000 –25.000 In most cases, hospitals that opt for an in-house service model (ppm) also rely on the provision of certain technical services from external companies, for example, the manufacturers of complex FIGURE I.17 31P NMR spectrum of the human brain showing Pi technologies (e.g. diagnostic imaging equipment). In such cases resonance. Inorganic scintillators 481 Insulation resistance Inorganic phosphate has a role in the body’s energy metabolism Further Reading: Knoll, G. F. 2000. Radiation Detection as a by-product of the dephosphorylation of ATP. Thus its concen- and Measurement, 3rd edn., John Wiley & Sons, New York, pp. tration, and the size of the resonance peak, increase in muscle 231–234. during exercise and in the brain during hypoxia/ischaemia. Input circuit Inorganic scintillators (General) The part of a device at which the (input) signal is (Nuclear Medicine) An inorganic scintillator is one out of two applied or where either a measuring sensor/transducer or a device groups of scintillators (inorganic and organic). The purpose of under test is connected. scintillators is to absorb high energy radiation and in the process The input circuit of electronic (signal measuring or sensing) produce scintillation light that can be measured using a photomul- devices is characterised by the input impedance and input signal tiplier tube. This detection setting is the most common in modern range. If the device under test is a passive device (not generat- emission imaging systems. Inorganic scintillation materials are ing any kind of power), the input circuit contains a power source solid crystals and the basic condition for scintillation comes from enabling measurement of electrical characteristics of the device characteristics in their lattice structure, individual atoms or mol- under test. ecules do not scintillate. In organic materials on the other hand the scintillation is a molecular property rather than an effect of Input curves on anode-cooling charts crystal structure. (Diagnostic Radiology) See Anode-cooling curve The scintillation property of an inorganic scintillating mate- rial depends on the energy states determined by the lattice struc- Input impedance ture. The electrons in such materials are only allowed in certain (General) See Input circuit discrete bands, the upper two bands are called valence band (the lower of the two) and conduction band (upper). In a pure crystal Input screen in fluoroscopy these two bands are separated by a gap of ‘forbidden’ energies. (Diagnostic Radiology) See Image intensifier The emitting of scintillation light is part of a two step process; (1) incident radiation excites electrons in the lower valence band up to the conduction band, (2) electrons release energy when they are Instantaneous dose rate (IDR) de-excited, thus sending out a scintillation photon. Some of these (Radiation Protection) The ionising radiation emitted from a transitions can be radiationless, i.e. not producing any photons. source may be continuous (e.g. from a radioactive source) or it Radiationless transitions are referred to as quenching. Quenching may be intermittent (e.g. from an electrically powered x-ray gen- leads to a non-linear response between deposited energy and the erator used for medical diagnosis or treatment). Each source will number of photons produced, although the problem is minimal be capable of delivering a radiation dose over a given period of relative to quenching in organic scintillators. A desired feature of time at a given distance dependent on the characteristics of the a crystal is a high light yield, namely a high number of photons source and any shielding or filtration present. per energy deposit. A crystal with high light yield provides a bet- In general, the instantaneous dose rate (IDR) is defined as the ter signal, hence better signal to noise ratio. radiation dose rate that may be

delivered at any given moment. In most pure crystals the energy gap between the two bands, For a radioactive source, this will be a continuous exposure and I and therefore the energy of the emitted photon, is too wide to simply related to the total activity present which will be decay- produce photons in the visible part of the spectrum. When using ing away over time according to the half-life for the radionuclide. photomultiplier tubes to strengthen the signal the ultimate scin- However, for electrically produced radiation the IDR may be tillation light is around 400 nm (blue light). A common approach extremely variable according to whether the radiation source is used to create photons in the visible part of the spectrum is to ‘on’ or ‘off’ and the settings used – the generating voltage, the add impurities to the lattice. These impurities create ‘islands’ of beam current and the exposure time (which could be from as little available energy states in the forbidden gap. The electrons will as tens of milliseconds and up to several minutes). de-excite via the impurity induced energy states and emit photons When determining the radiation protection shielding and other with lower energies. features of a radiation facility to ensure the safety of staff and the A desired crystal feature is that the half-life of each excited public, it is appropriate as a first step to consider the IDR that the energy state is short. A long decay time will lead to dead time radiation source is capable of delivering, and then to consider the losses (see separate article) because the chance of two signals way in which the source is used to determine the radiation dose overlapping increases. Typical decay times are 50–500 ns. that any person may receive over a longer period of time – by The use of impurities also introduces an unwanted effect time-averaging the dose rate to which a person may be exposed. called phosphorescence or afterglow. Some electrons are ‘caught’ For most radiation sources, it will be this time-averaged dose in energy states where further de-excitation is forbidden. The rate that will be of more importance in determining the radia- only way to de-excite to the ground state is to first excite to a tion protection design features and working practices that may higher-lying energy state. The photons produced in such cases are be required to reduce the total dose received to acceptable and delayed in time, which has a degenerative effect on the signal. optimised levels. Crystals with impurities have no extensive self-absorption Related Articles: Radiation dose, Time-averaged dose rate of the emitted radiation as is the case with pure crystals. The (TADR), Time-averaged dose rate (TADR2000) impurities are spread out over the lattice and work as emitter and absorber centres, so the probability of re-absorption is lower than Insulation resistance in a pure crystal where all atoms can absorb the scintillation light. (General) The resistance between two conducting parts (elec- Related Articles: NaI (Tl) detector crystal, Scintillators, trodes) that are under normal conditions are insulated from each Phosphorescence, Light yield in scintillation detectors, Bismuth other. Insulation resistance is of the order of megaohms [MΩ]. germanate (BGO) Proper insulation resistance is important for patient and staff Integral dose 482 Integrating dosimeter safety in all medical devices, but especially in those which oper- The introduction of three dimensional treatment planning sys- ate with high voltage (x-ray, CT, US). tems has permitted the introduction of an alternative method of displaying the results of dose calculation by the dose volume his- Integral dose tograms (DVH). The DVHs easily permit comparison of different (Radiotherapy) One way of comparing dose distributions for dif- radiotherapy plans. ferent quality beams is to calculate the integral dose for a given Further Reading: Mayneord, W. V. 1942. The measurement tumour dose. The integral dose is a measure of the total energy of radiation for medical purposes. Proc. Phys. Soc. 54:405. absorbed in the treated volume and it is desirable to keep it as low as reasonably possible. Except the case of large volume Integrated backscatter irradiation integral dose is not a limiting factor in radiotherapy, (Ultrasound) Integrated backscatter is the name for a form of tis- mainly in the modern techniques. If a mass of tissue receives sue characterisation using ultrasound. The measure is the ratio of a uniform dose, then the integral dose is simply the product of the power spectrum of the received signal divided by the power mass and dose. However, in practice, the absorbed dose in the spectrum of the signal from a standard reflector in the focal plane tissue is non-uniform so a mathematical formula is required to of the transducer. Through Parseval’s theorem, the division can calculate it. also be performed in the time domain, using the square of the For a single beam of photons an approximate method accurate envelope of the time domain signal. Integrated backscatter can enough was introduced by Mayneord who formulated the follow- also be compensated for attenuation. ing approximate expression: The application for integrated backscatter has mainly been in cardiac application, where it has gained some success. Due to d the structure and movement of the heart, the backscattered sig- S = 1. ( - -0.693 / )æ 2.88 1/2 1/2 ö 44D0 Ad1/2 1 e d d ç1+ ÷ è SSD ø nal from heart tissue will vary over the cardiac cycle. Where the sound beam is parallel to the muscle fibres of the heart, little sig- where nal is received, whereas when the beam is perpendicular to the Σ is the integral dose fibres, the signal power is increased. The opposite occurs to angu- D0 is the delivered dose lar dependence of frequency-averaged attenuation. The timing of A is the field area these variations, as well as the relative change of integrated back- d is the total thickness of the patient in the path of the beam scatter and attenuation, relate to various disorders like infarcts, d1/2 is the depth of 50% isodose ischaemia and myopathy. SSD is the source-surface distance and the expression inside the brackets takes into account the geometric diver- Integrated parallel acquisition technique (iPAT) gence of the beam (Magnetic Resonance) See iPAT (integrated parallel acquisition technique) The unit for integral dose is kilogram gray or simply joule. The integral dose depends on the field arrangement and the Integrating dosimeter beam quality. In Figure I.18 the integral dose as a function of (Radiation Protection) The integrating dosimeter measures I the beam quality is shown for a tumour dose of 10 Gy at a depth the dose of ionising radiation during a known period of time. of 12.5 cm in a 25 cm thick patient treated with parallel opposed Integrating dosimeters play a very important role in personnel beams, field size 10 cm × 10 cm at an SSD = 100 cm. The curve monitoring as well as in radiation protection of the patient. shows that the higher the photon energy the lower the integral There are active and passive integrating dosimeters. Typical dose. active integrating dosimeters are gas-filled detectors, for exam- ple small size ionisation chambers called ‘pocket chambers’ (Figures I.19 and I.20). The passive integrating dosimeters are of different kinds: 3400 photographic films, solid state nuclear track detectors, 3300 3200 3100 3000 2900 2800 60Co 10 20 30 Beam quality (MV) FIGURE I.18 Integral dose as a function of the beam quality for a tumour dose of 10 Gy at a depth of 12.5 cm in a 25 cm thick patient treated with parallel opposed beams, field size 10 cm × 10 cm at an SSD = 100 cm. FIGURE I.19 Pocket dosimeter with a charger (example). Integral dose (kg cGy) Integrating the healthcare enterprise (IHE) 483 Intelligent workstation implement and enable care providers to use information more effectively (Figure I.21). IHE enhances the quality of patient care, resulting in the fol- lowing benefits: • Safety through the reduction of medical errors • Savings through lower implementation costs and more efficient workflow • Satisfaction through better informed medical decisions and faster results for both patient and physician Hyperlinks: www .ihe .net FIGURE I.20 Pocket dosimeter, exposure scale. Intelligent workstation (Diagnostic Radiology) An ‘intelligent workstation’ refers to a reporting workstation that incorporates computer-aided diagnosis thermoluminescent dosimeters, activation foils. See related arti- (CAD) to act as a ‘second opinion’ to a reporting radiologist by cles for examples. highlighting areas in a medical image that may be of diagnostic Related Articles: Dosimeter, Film badge, Finger ring significance. CAD involves computer algorithms for image pro- dosimeter, Geiger–Müller (GM) counters, Ionisation chamber, cessing, image feature analysis and data classification using, for Thermoluminescent dosimeter (TLD) example, artificial neural networks, and is commonly referred to Further Reading: Knoll, G. F. 2000. Radiation Detection and as ‘artificial intelligence’. Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. Initial intelligent workstations consisted of a high-speed 139–145. computer, x-ray film digitiser, image archive and hard and soft copy outputs. Multiple film radiographs from the same patient Integrating the healthcare enterprise (IHE) would be supplied to the system, where they were digitised and (Diagnostic Radiology) IHE (integrated healthcare enter- displayed. CAD software would then highlight areas of interest prise) is an initiative by healthcare professionals and the indus- in the image. Radiologists could then assess current and previ- try to improve the way computer systems in healthcare share ous images with areas of diagnostic interest highlighted by the information. system. Systems were used to, for example, highlight suspected IHE promotes the coordinated use of established standards pulmonary nodules in chest radiographs and detect microcalcifi- such as DICOM and HL7 to address specific clinical need in sup- cations in mammograms. port of optimal patient care. Systems developed in accordance The introduction of digital image acquisition and picture with IHE communicate with one another better, are easier to archiving and communication systems (PACS) has greatly I IHE connectathon Product IHE results integration statement Standards Implementation development: projects HL7, ISO, IHE IHE DICOM, etc. Technical connectathon demonstration framework Product Easy to with IHE integrate products Standards Integration profile RFP Requirements + specifications Testing Interoperability FIGURE I.21 The IHE process. Intense pulsed light source (IPL) 484 Intensity modulated proton therapy (IMPT) improved the ability to incorporate CAD into regular image if the film were exposed directly by the x-radiation. At least two reporting and has already become a part of routine clinical work factors contribute to this effect. The intensifying screen is thicker for the detection of breast cancer in mammography. and a more efficient x-ray absorber than the film emulsion and the Related Articles: Workstation, PACS (Picture archiving and increase in photon concentration produced by the x-ray to light communication systems), CAD (Computer-aided diagnosis), conversion is more effective in producing film exposure. Artificial neural networks Until the 1970s calcium tungstate was the typical fluorescent Further Readings: Giger, M. L. et al. 1993. An ‘intelligent material in radiographic intensifying screens. It has been replaced workstation’ for computer-aided diagnosis. RadioGraphics by rare earth fluorescent materials containing lanthanum or gado- 13:647–656; Nishikawa, R. N. et al. 1995. Initial experience with linium. Contemporary intensifying screens use yttrium. Screen- a prototype clinical intelligent mammography workstation for film systems are now quickly replaced by computed radiography computer-aided diagnosis. Proc. SPIE 2434, Medical Imaging systems. 1995: Image Processing (12 May 1995); Shiraishi, J. et al. 2011. Related Article: Screen film Computer-aided diagnosis and artificial intelligence in clinical imaging. Semin. Nucl. Med. 41(6):449–462. Intensifying screen(s), rare earth (Diagnostic Radiology) See Rare earth screen Intense pulsed light source (IPL) (Non-Ionising Radiation) An intense pulsed light source is a non- Intensity coherent optical radiation source with output across several wave- (Ultrasound) Intensity is defined as the energy flux crossing a lengths (polychromatic). A high voltage power supply is applied -surface per second, i.e. it is the power per unit area. It has units to a quartz tube filled with a noble gas such as xenon or krypton of J/m2/s or W/m2. (a flashlamp). Following the initial ‘flash’, electrons (current) can Its value at any point is proportional to the pressure amplitude flow through the gas causing ionisation. Ions recombine with lost squared of the sound wave: electrons and lose energy in the form of light. The light produced is filtered to provide the required wave- i.e. I = 2

p2 /rc = 2 p2 /Z assumingplanewaves length range for treatment and is usually delivered to the treat- ment area using a handpiece. The handpiece usually contains the where flashlamp, filter and a lens or waveguide. Other beam delivery ρ is density of medium methods such as optical fibres may also be used. The effects on c is speed of sound the skin are similar to those seen when using lasers. IPLs are gen- Z is the acoustic impedance erally used in the cosmetic sector for hair removal, photo-rejuve- nation and treatment of vascular lesions. For a diagnostic ultrasound beam the acoustic intensity will Under IEC 62471:2006 Photobiological safety of lamps and vary along the length of the beam as the beam cross-section lamp systems, an IPL source is classified as a Risk Group 3 source. changes shape due to focusing, and energy is attenuated due to Under the Control of Artificial Optical Radiation at Work absorption and scattering. Intensity is therefore usually greatest Regulations, 2010 an in-depth risk assessment must be carried I nearer to the transducer and at transmit focal points, where all the out for Risk Group 3 equipment. energy is passing through a small cross-sectional area. Intensities Related Articles: Hazard value, IPL are also usually greatest for non-scanned modes such as m-mode Further Readings: A Non-Binding Guide to the Artificial and pulsed wave Doppler, as the beam is then stationary within the Optical Radiation Directive 2006/25/EC, Radiation Protection tissue, and when pulse-average intensities are higher such as pulsed Division, Health Protection Agency; Medicines and Healthcare wave Doppler and colour Doppler where longer pulses are used. products Regulatory Agency, Lasers, intense light source sys- Heating of tissue is proportional to the intensity making inten- tems and LEDs – guidance for safe use in medical, surgical, den- sity an important parameter to know when considering the safety tal and aesthetic practices, Crown copyright, September 2015; of ultrasound and the likelihood of thermal damage. International Electrotechnical Commission: IEC 62471:2006 In order to accurately measure intensity it is necessary to make Photobiological safety of lamps and lamp systems. IEC 2006. measurements over all frequencies that may be present, including the high harmonic frequencies that may exist due to non-linear Intensification factor (IF) propagation within the beam. (Diagnostic Radiology) Intensification factor is a characteristic It is often useful to further define how intensity is measured by of a specific intensifying screen and film combination. It is the considering spatial peak or average intensities and instantaneous ratio of exposure directly to a film compared to the exposure to or time average intensities. an intensifying screen-film combination required to produce the Related Articles: Pulse average intensity, Spatial peak inten- same optical density. IF has very little meaning or application in sity, Spatial average intensity, Time average intensity, Temporal modern radiography because direct film exposure is not a useful peak intensity reference. Intensity modulated proton therapy (IMPT) Intensifying screen (Radiotherapy) Spot scanning can be used to deliver intensity (Diagnostic Radiology) Intensifying screens (or intensifying modulated proton therapy (IMPT), which is the equivalent of phosphors) are sheets of fluorescent materials placed in contact intensity modulated radiotherapy (IMRT) in high energy x-ray with the x-ray film in radiographic cassettes to perform two func- radiotherapy. In IMPT, multiple fields are used. Each field has an tions. It absorbs the x-radiation and converts a fraction of the inhomogeneous proton fluence distribution. A homogenous dose absorbed energy to light. The light exposes the film. The advan- distribution can be delivered to the target volume by combining tage is that an image can be created with much less exposure than multiple inhomogeneous proton fields (Khan and Gibbons, 2014). Intensity-modulated radiotherapy (IMRT) 485 Interactive implant technique Relative to single field uniform dose (SFUD) proton therapy, In ideal circumstances, if there is uniform insonation of the IMPT can achieve greater target conformality and normal tissue vessel, the mean velocity is obtained from the intensity weighted sparing. One disadvantage of IMPT is that it has the potential to mean of the Doppler spectrum, defined as be especially sensitive to range uncertainties (McGowan et al., 2013). An uncertainty of a few millimetres could lead to under- ò P ( f ) f df dosage in the target and/or over-dosage of organs at risk. f (t ) = f Related Articles: Intensity-modulated radiotherapy (IMRT), ò P ( f )df Single field uniform dose (SFUD), Multiple field optimisation f (MFO) Further Readings: Khan, F. M. and J. P. Gibbons. 2014. where P( f) is the Doppler power spectrum. Khan’s the Physics of Radiation Therapy, 5th edn., Wolters The intensity weighted mean can be obtained by analogue Kluwer Health, Philadelphia, PA; McGowan, S. E., N. G. Burnet means or, in modern systems, by digital analysis of the spectrum and A. J. Lomax. 2013. Review article: Treatment planning opti- analyser. misation in proton therapy. Br. J. Radiol. 86:20120288; If the sample volume does not encompass the entire vessel or if there is non-uniform insonation then the calculated intensity weighted mean will not accurately represent the mean velocity. Intensity-modulated radiotherapy (IMRT) Non-axial flow can also lead to errors. The time averaged mean (Radiotherapy) Intensity-modulated radiotherapy (IMRT) is a set velocity (TAMV) (Figure I.22) shows the mean of this value over of techniques of varying the intensity across the radiation field five cardiac cycles. to deliver complex 3D dose distributions in external beam radio- therapy. It is found to be particularly useful in creating dose dis- tributions with concavities, which are needed when the target is in Interaction close proximity to a dose limiting critical organ. (Radiation Protection) A reaction between radiation and matter. Conventional radiotherapy uses a set of beams with fixed Interactions may be broadly divided into particle-particle reac- intensity profiles delivered in a cross fire effect to achieve a tions, atomic interactions (where the radiation interacts with high dose to the target with maximal sparing of adjacent tis- atom-bound electrons) and sub-atomic interactions (where the sues. In IMRT the intensity profile of these beams is shaped radiation interacts with the atom’s nucleus). such that the combined effect achieves the desired dose distri- bution. The added complexity generated by the extra degrees of Interactive implant technique freedom achievable with beam profile shaping means that the (Radiotherapy, Brachytherapy) conventional manual, forward or interactive planning approach Source Handling and Loading: Brachytherapy sources must is generally not the best manner to plan the treatment and an be handled and loaded into the applicators for treatment, and automatic approach called inverse planning, or inverse opti- many methods have been used over the time. These methods have misation is used. This is a mathematical method of generating been developed primarily to reduce the dose to the personnel, but the treatment plan, in which a prescription is specified by the also to improve the quality of the treatment itself. treatment planner in terms of doses to tumour and dose limits Interactive Implant Technique: Example: Ultrasound-guided to critical structures. The computer then plans the beam distri- interstitial interactive permanent implant of Iodine-125 seeds, an I butions that produce a dose distribution that best matches the intra-operative procedure – prostate cancer. prescription. Permanent prostate implants are often performed with man- Delivery of IMRT is most often achieved on a standard linac ual loading and manual afterloading techniques, as low energy using a multileaf collimator, either as a set of static fields of dif- sources are used. During these procedures, dose to staff from fering shape from each beam direction or by scanning the leaves the sources themselves is very low (Figures I.23 through I.25). If across the field during irradiation in the dynamic MLC technique. Other delivery systems for IMRT exist including intensity-modu- lated arc therapy, tomotherapy and robotic radiotherapy. Abbreviations: IMRT = Intensity-modulated radiotherapy. Related Articles: Conformal radiotherapy, Multileaf collima- tor, Inverse planning, Forward planning, Interactive planning, Tomotherapy, Robotic linacs, Gamma knife Further Reading: Webb, S. 2000. Intensity Modulated Radiation Therapy, Taylor & Francis Group, London, UK. Intensity reflection coefficient (Ultrasound) See Reflection coefficient Intensity transmission coefficient (Ultrasound) See Transmission coefficient Intensity weighted mean (Ultrasound) In many clinical applications, it is useful to be able to determine the mean velocity in a vessel. In combination with a FIGURE I.22 Measurement of volume flow through a vessel. The red measurement of cross-sectional area, mean velocity will give the line in the sonogram shows the calculated intensity weighted mean veloc- volume flow through a vessel. ity throughout the period measured. Interactive implant technique 486 Interactive implant technique fluoroscopy is used during the implant procedure, the dose to staff 6. Verification of source strength (dosimetry equipment – comes mainly from the fluoroscopy. electrometer and well-type chamber, insert suitable for The patient is sleeping during the whole procedure (requires the seeds used) anaesthesia equipment), and the procedure is performed in an 7. Preparation of needles and seeds/strands (seed handling operating theatre. The interactive implant procedure consists of equipment, implantation needles) 8. Interactive implantation (ultrasound unit with bi-plane 1. Patient positioning (suitable treatment table, including rectal probe, stepper for the rectal probe, template to leg supports) guide the needles, fluoroscopy unit, dedicated interac- 2. Requirement: identical patient position during the tive treatment planning system) whole procedure! 9. Source positions are adjusted in the TPS according to 3. Image collection ultrasound (US) unit with bi-plane US and fluoroscopy, the dose distribution is interac- rectal probe, stepper for the rectal probe tively updated 4. Definition of target and organs-at-risk (dedicated treat- a. 50–100 sources used ment planning system, TPS) b. Inter-source effects; not included in the dose 5. Treatment planning – and plan acceptance calculations! 10. Book-keeping of sources – total number accounted for a. Implanted sources counted using fluoroscopy I FIGURE I.23 Sagittal US-section; Foley catheter in the bladder, with FIGURE I.25A Transverse US-image; one needle inserted in position contrast in the balloon. E5, no disturbing gas. FIGURE I.24 Sagittal and transverse US-images, note the ‘air’ problem which destroys the image information. The transverse image shows the target (the prostate) (big circle), the urethra (small circle), the rectum (bottom arc) and the pubic arch (upper arc). Also shown is an overlay of the template. One needle is inserted, at position E5. The needle echo, indicated by an arrow, is seen despite the image degradation. Interactive planning 487 Interleaved FIGURE I.26 Interference from two point sources emitting FIGURE I.25B Fluoroscopy image; two needles inserted, E5 and I5, continuously. both containing three seeds, the seeds are still in needles. The US-probe in the rectum and the contrast-filled Foley catheter balloon in the bladder are seen between the two needles. depends on the difference in distance from the observation point to the respective sources. Related Articles: Brachytherapy, Source loading in brachy- Interlacing monitors therapy, Temporary implant, Permanent implant, Interstitial (Diagnostic Radiology) See Acquisition modes for digital image brachytherapy Interleaved (Magnetic Resonance) In MRI, the data collection can be inter- Interactive planning leaved in four different ways. (Radiotherapy) Radiotherapy treatment planning involves Interleaved k-space Coverage: The k-space can be acquired choosing the beam directions, field shapes and intensities for on one or more shots. When using a multi-shot technique, the the patient’s beam delivery. This is usually done using a CT lines are often sampled interleaved. The first acquisition collects a scan of the patient and a model of dose deposition in the patient series of lines in k-space and the second collects lines in between by the beam. A common approach to this is the interactive, or the first lines (see Figure I.27). Motion between shots has to be trial and error approach, in which beam parameters are tried avoided. by the planner and the resulting dose distribution evaluated in Interleaved Slice Acquisition: When different slices of a scan terms of target coverage and dose to non-target tissues. Often I overlap with each other a Crosstalk artefact can appear. This is dose volume histograms are used as part of the evaluation of caused by a slice profile that is not ideal due to constraints of the the treatment plan. This is also sometimes referred to as for- ward planning. The alternative to this is inverse radiotherapy planning. k Abbreviation: CT = Computed tomography. y Related Articles: Treatment planning, Inverse radiotherapy planning, Dose volume histogram Interface (Ultrasound) In ultrasound contexts the word interface is used when discussing borders between two materials with different

acoustic impedances, for example the transducer-skin interface and blood-myocardium interface. Related Articles: Reflection coefficient, Acoustic impedance, kx Reflection Interference (Ultrasound) Two waves that travel together can, dependent on their respective phase, be observed as a wave that is the sum of the two waves’ individual amplitudes (constructive interference), or add up to no apparent wave motion at all if the amplitudes 1st shot are equal and the phases are opposite (destructive interference). 2nd shot In Figure I.26 two point sources emit continuous waves, and in certain directions the waves appear to be in phase, whereas in oth- FIGURE I.27 Interleaved k-space coverage for a collection using two ers, to be out of phase. As can be deduced from Figure I.26, this shots. Interlock; Interlocking device 488 Internal conversion measurement technology. To avoid crosstalk the slice gap can be ensure that staff are not exposed to high doses, and as a fall-back increased or the slice order may be interleaved. With interleaved to limit the consequences of any malfunction of the machine. slices the slices are not collected one by one in a row but in an Interlocks are examples of active engineered controls, which order that maximises the distance between two slice collected monitor a changing situation and can trigger a safety action, either after each other, see Figure I.28. Interleaved slice collection can electronically, or mechanically. produce larger mean intensity differences between the slices (e.g. Common examples of interlocks are as follows: due to the brain size) and inflow effects may change local signal strength. Multi-Slice Interleaved k-space: By combining the two meth- • The ‘last man out’ button and door interlock. The ods given earlier a third method is constructed. A series of lines machine will not irradiate unless the ‘last man out’ but- can be collected in each slice before returning to the first slice to ton has been pressed and the door is closed. The posi- acquire the second interleave, see Figure I.29. Since the acquisi- tion of the ‘last man out’ button is such that it gives a tion time of each separate slice is long the risk of severe motion clear view for the operator to ensure that no other per- artefacts is increasing. son remains in the room, other than the patient. If the Interleaving Acquisitions: This may refer to alternating door is opened during irradiation, then the beam will application of different pulse sequences within a single acquisi- immediately be turned off. tion (e.g. inversion recovery and spin echo in some T1 measure- • Interlocks are used extensively with the computer con- ment techniques), or to alternating MRI data acquisition with trol of MLCs. For example, the beam will be inter- some other technique such as ultrasound or EEG. locked if a misplaced MLC leaf is detected, or if the Related Articles: Crosstalk, k-space, Motion artefacts MLCs are unintentionally used during electron mode. • Many interlocks will be associated with the ion cham- Interlock; Interlocking device bers that monitor the beam energy, flatness and symme- (Radiotherapy) The design of radiotherapy treatment machines try. If any of these measured quantities drift out of the and rooms includes many interlock safety features, in order to tolerance levels, then the interlock will be activated and irradiation will cease. Additionally, the beam is inter- locked if the readings from the two ion chambers differ significantly. 1 3 2 4 • There are also interlocks for the correct selection of target, filter, scattering foil to match that at the control panel. Abbreviations: MLC = Multileaf collimator. Interlocking mechanism (Radiotherapy) See Interlock; Interlocking device I Internal beam irradiation (Nuclear Medicine) Cyclotrons are used to produce radionuclides by accelerating charged particles and then allowing them to hit a target. One way to irradiate the target is to insert it into the par- ticle beam. This technique is called internal beam irradiation. Using this technique allows most of the accelerated particles to hit FIGURE I.28 In interleaved slice selection the slices are collected in a the target. The second alternative is to extract the particle beam way that increases the distance between two slices collected after each by using either a stripping foil or an electrostatic deflector. This is other. called external beam irradiation. In general, external beam irra- diation is the preferable method. Related Articles: Cyclotron, External beam irradiation, 1 4 Stripping foil Further Reading: Cherry, S. R., J. A. Sorenson and M. E. Slice 1 Phelps. 2012. Physics in Nuclear Medicine, 4th edn., Elsevier Saunders, Philadelphia, PA, Chapter 5. 2 5 Internal conversion (Radiation Protection) If an atomic nucleus is in an excited state Slice 2 A Z N * it releases the excess energy Eexc either by photon emission (gamma radiation) or internal conversion. During internal con- 3 6 version the excess energy Eexc, which would have been released by emission of gamma radiation, is instead absorbed by a bound kx electron, usually from a K or L atomic shell, and then that electron Slice 3 is ejected from the atom with a kinetic energy Eke equal to ky Eke = Eexc - Eb FIGURE I.29 Interleaved k-space coverage for three slices using two shots in each slice. where Eb is the binding energy of the electron. Internal margin 489 Internal photoelectric effect e– Conversion electron PTV x-ray ITV CTV K A GTV L Z N* OAR FIGURE I.31 Target volumes as described in ICRU 62 (Podgorsak). A FIGURE I.30 Scheme of internal conversion process for nuclei Z N * and emission of a characteristic x-ray. The IM incorporates both intra-fraction errors such as that due to respiration, and inter-fraction errors such as that due to weight gain/loss or digestive system changes. The latter inter-fraction The vacancy created on the atomic shell after an internal con- error can be compensated to an extent by the use of portal imag- version is filled by an electron from the higher shell (Figure I.30) ing during each fraction. with the subsequent emission of a characteristic x-ray. The set-up margin (SM) is also required to account for errors The output of an internal conversion process can be estimated in patient positioning and alignment of treatment beams, and the using a coefficient α, defined as the proportion of nuclear transi- ITV plus this additional margin defines the extent of the PTV. tions which result in internal conversion, and can be compared to The total required margin (SM + IM) can be quantified from the the proportion resulting in gamma emissions. This coefficient can standard deviations (σ) of the associated probability distribu- be further refined to describe the rate of K-shell internal conver- tions. If the error distributions can be treated as Gaussian and sion – i.e. αK; similarly the rate for L-shell internal conversion – αL independent, then their standard deviations should be combined and for M-shell internal conversion – αM. in quadrature (i.e. sum of variances), with a multiplication factor according to the level of confidence, as shown in Equation I.2. In other words, the total margin would be proportional to the EXAMPLE: combined standard deviation: In the isomeric decay of 99mTc to 99Tc, about 9% of the Total margin a s2 SM + s2 IM (I.2) nuclear transitions result in K-shell internal conversion, I where s2 1.1% in L-shell, and 0.3% in M-shell. Then the total inter- IM = s2 IMi + s2 nter IMi intra nal conversion coefficient is equal to 10.4%, whilst the Equation I.2 shows combining margins for set up error and physi- remaining 89.6% of nuclear transitions result in gamma ological (internal) error. radiation emission. However this approach is not commonly used in practice to calculate margins. A more practical method is to decompose the margin into random and systematic components as explained in Related Articles: Gamma rays, Isomeric transition (IT) Set-up error. Further Readings: Graham, D. T. and P. Cloke. 2003. Abbreviations: CTV = Clinical target volume, GTV = Gross Principles of Radiological Physics, 4th edn., Elsevier Science tumour volume, ITV = Internal target volume, OAR = Organ at Limited, Edinburgh, UK, p. 257; Hobbie, R. K. 1997. Intermediate risk, PTV = Planning target volume and SM = Set up margin. Physics for Medicine and Biology, 3rd edn., Springer-Verlag, New Related Articles: IM internal margin, Set up error, Portal York, pp. 462–463; Stabin, M. G. 2008. Radiation Protection imaging, Gross tumour volume (GTV), Clinical target volume and Dosimetry: An Introduction to Health Physics, Springer (CTV), Planning target volume (PTV), ICRU Science+Business Media, LLC, New York, p. 28. Further Readings: ICRU (International Commission on Radiation Units and Measurements). 1993. Prescribing, Reporting Internal margin and Recording Photon Beam Therapy, ICRU 50, Washington, (Radiotherapy) A margin needs to be added to the clinical tar- DC; ICRU (International Commission on Radiation Units and get volume (CTV) to form the planning target volume (PTV) to Measurements). 1999. Prescribing, Recording and Reporting account for any positional errors from the planning information. Photon Beam Therapy (Supplement to ICRU Report 50), ICRU The ICRU Report 62 divided this margin into the set-up margin Report 62, ICRU, Washington, DC; Podgorsak, E. B. 2005. Review (SM) and internal margin (IM), in order to separate out the con- of Radiation Oncology Physics: A Handbook for Teachers and tributory sources of positional error into physiological error and Students, International Atomic Energy Agency, Vienna, Austria. set up error, respectively. The IM compensates for physiological variation of the size and shape of the volume. It defines the differ- Internal photoelectric effect ence between the CTV and the internal target volume (ITV), as (Radiation Protection) An interaction between photonic radiation shown in Figure I.31. (ionising or non-ionising) and a material in which the absorption Internal radiation dosimetry 490 International Commission on Non-Ionising Radiation of a photon in a material results in the excitation of an electron purposes to relate image-guided techniques to traditional plan- from the valence band to the conduction band. This is of particu- ning methods. lar significance in the design of semiconductor radiation detec- The ICRU rectum and bladder point doses whose position can tors, for example used in modern digital radiology. be determined from orthogonal radiographs often do not reflect Related Articles: Photoelectric effect, External photoelectric the true maximum doses, as has been shown from studies using effect 3D image based treatment planning. This is especially true for bladder where the dose determined using the ICRU reference Internal radiation dosimetry point dose can significantly underestimate the maximum bladder (Radiation Protection) Internal radiation dosimetry employs dose value. A discussion of doses to organs at risk in relation to biokinetic models and Monte Carlo mathematical techniques to different side effects, maximum doses, doses to 1 cm3, and 2 cm3 determine the effective dose to a standard sized adult, and to chil- volumes, etc. can be found in the following references. dren at various ages, from the intended or accidental adminis- As an example, Figure I.32 shows two orthogonal radiographs tration, ingestion or inhalation of a wide range of radionuclides used to determine bladder and rectum ICRU points (slightly in their common chemical forms. This includes radiopharma- modified). A picture of the ring applicator itself and a stan- ceuticals used in nuclear medicine either for therapeutic or diag- dard dose distribution are included in the article Intracavitary nostic purposes. An example of data available for use in nuclear brachytherapy. medicine is the medical internal radiation dose (MIRD) tables and software. The biokinetic models the path of the radioactivity 1. Ring applicator including plastic spacers (not visible on through the human body, describing how long the radioactivity the images) remains in each organ or tissue. Knowledge of the total activity a. Diameter ring including spacer = 3.4 cm administered, and the biokinetic model, leads to the calculation of b. Intrauterine probe length = 4 cm the absorbed dose to each organ over the time the activity remains in that organ (the committed absorbed dose), and thus to the effec- 2. 7 mL contrast in Foley catheter balloon in bladder (Bl) tive dose over the time that the activity remains in the body (the 3. Special indicator with two rows of lead-shot in rectum committed effective dose). (a, b) Related Articles: Effective dose, Committed dose 4. Silver marker in cervix (elongated – for verification of appl. pos. between fractions) Internal reference point 5. Marker wires indicating source stop (dwell)

positions (Radiotherapy, Brachytherapy) inserted in both probe and ring Reference Points in Brachytherapy: In classical brachyther- 6. Calculations of total radiobiological effect (external apy, reference points were defined to specify dose to target and beam radiation and brachytherapy) for organs at risk to calculate dose to organs at risk. Reference points were often (dose restrictions) and target are made defined in relation to the applicator(s), e.g. point A used to specify dose to the target in the Manchester system for treatment of cervix Abbreviations: ICRU = International Commission on carcinoma. Radiographs were used to define reference points, and Radiation Units and Measurements. I in the ICRU Report 38 standard reference points for rectum and Related Articles: Intracavitary brachytherapy, Volumetric bladder were defined and recommended for reporting in gynaeco- prescribing – brachytherapy logical brachytherapy. Further Readings: Haie-Meder, C. et al. 2005. The ‘simple’ ICRU bladder and rectum reference points, based Recommendations from gynaecological (GYN) GEC-ESTRO on orthogonal radiographs, have been used extensively to charac- Working Group (I): Concepts and terms in 3D image based 3D terise BT for cervix cancer in terms of maximum doses to these treatment planning in cervix cancer brachytherapy with emphasis organs, in spite of their well known shortcomings. These ICRU on MRI assessment of GTV and CTV. Radiother. Oncol. 74:235– reference points, as well as point A, are still used for reporting 245; ICRU (International Commission on Radiation Units and a b Sin βL βL b a FIGURE I.32 HDR remote afterloading technique with ring applicator, using orthogonal x-ray films. International Commission on Radiological Protection 491 International Commission Measurements). 1985. Dose and volume specification for report- International Commission on Non-Ionising ing intracavitary gynaecology, ICRU Report 38, Washington, DC; Radiation Protection (ICNIRP) Lang, S. et al. 2006. Intercomparison of treatment concepts for (General) The principal focus and aim of ICNIRP is to dissemi- MR image assisted brachytherapy of cervical carcinoma based on nate information and give advice on health hazard regarding the GYN GEC-ESTRO recommendations. Radiother. Oncol. 78:185– exposure to non-ionising radiation; including the optical radiation 193; Pötter, R. et al. 2006. Recommendations from gynaecologi- (ultraviolet, visible and infrared, etc.), time-varying electric and cal (GYN) GEC ESTRO working group (II): Concepts and terms magnetic fields, radiofrequency radiation and ultrasound. in 3D image-based treatment planning in cervix cancer brachy- Evaluations of the risk associated with the non-ionising radia- therapy – 3D dose volume parameters and aspects of 3D image- tion are usually carried out in collaboration with the World Health based anatomy, radiation physics, radiobiology. Radiother. Oncol. Organisation and the results are published as safety guidelines and 78:67–77. when feasible include exposure limits. When the workers’ safety is involved, the collaboration is extended to the International Internally deposited radionuclide Labour Organisation (ILO). The activity of ICNIRP is based on (Nuclear Medicine) An internally deposited radionuclide is a wide scientific expertise including medicine, epidemiology, biol- radionuclide which is deposited in a dispersed form within the ogy, physics, dosimetry, etc. Therefore, in addition to the WHO body. Examples include solutions, gases, dusts, and suspensions. and ILO, it is essential the partnership with the other related These substances can enter the body by inhalation, ingestion, international organisations such as the International Radiation injection or percutaneous absorption. Protection Association (IRPA), the professional representative body for radiation professionals world-wide, the Institute for International Atomic Energy Agency (IAEA) Electrical and Electronic Engineers (IEEE), the International (General) The International Atomic Energy Agency (IAEA) was Electrotechnical Commission (IEC), the European Commission created in 1957 (‘Atoms for Peace’). It is an international organ- (EC) and many others. isation related to the United Nation (UN) as the worldwide centre ICNIRP is legally registered in Germany and is a non-profit for co-operation for the peaceful use of atomic energy. Medical organisation. The modest income comes from IRPA, the German applications are also included. In fact in its mandate as it is stated Environment Ministry, other governments and other sources with in article II of the IAEA statute is mentioned that: the agency exemption of industries. Some income is provided by the sale of shall accelerate and enlarge the contribution of atomic energy to the publications. Many reports can be downloaded for free. health (recently this has been enlarged including also x-rays). The structure of ICNIRP consists of a main commission and Typically, the IAEA is working in partnership with its Member several standing committees. Each standing committee has its States. The relation with the UN is regulated by a special agree- own work carried out in agreement with the main commission. ment. The Agency reports annually to the UN General Assembly Independent experts are called in to participate in the work of and, when appropriate, to the Security Council. the different committees; they do not represent (either) their own The Headquarter of the IAEA is based in Vienna, Austria, at country or their institute. They cannot be employed by industries. the Vienna International Centre. Operational liaisons are located The expert work is given voluntarily. in other countries, namely Switzerland, the United States, Canada Hyperlinks: www .icnirp .de; www .earthprint .com and Japan. There are also research centres and scientific laborato- Related Articles: Action spectra, AORD, Effective exposure I ries run or supported by the agency, located in Austria, Monaco Further Readings: Coleman, A., F. Fedele, M. Khazova, P. and Italy. In 2007 the regular budget of the IAEA was of 283, Freeman and R. Sarkany. 2010. A survey of the optical hazards 611, 000 Euro (plus 80 million USD of voluntary contributions to associated with hospital light sources with reference to the Control the Technical Co-operation Fund), with a staff of 2200 multidisci- of Artificial Optical Radiation at Work Regulations 2010. J. plinary professional and support staff coming from 90 countries. Radiol. Prot. 30(3):469; ICNIRP. 2000. Revision of the guidelines Policy making bodies plan the work programme and the budget. on limits of exposure to laser radiation of wavelengths between Within the aims of the agency, the programmes of work are 400 nm and 1.4 µm. Health Phys. 79(4):431–440; ICNIRP. 2004. tailored in order to support the various member states in their Guidelines on limits of exposure to ultraviolet radiation of wave- needs. In general there are three main areas of work: safety and lengths between 180 nm and 400 nm (incoherent optical radia- security; science and technology; safeguard and verifications. tion). Health Phys. 87(2):171–186; ICNIRP. 2013. Guidelines on Medical applications fall under the first two topics. limits of exposure to incoherent visible and infrared radiation. The agency is structured in departments. The activities are Health Phys. 105(1):74–91; ICNIRP. 2013. Guidelines on limits grouped as (1) technical cooperation (mainly cooperative projects of exposure to laser radiation of wavelengths between 180 nm and in developing countries, providing training, specialised equip- 1,000 µm. Health Phys. 105(3):271–295; ICNIRP. 2016. A closer ment and general support); (2) research and development (support look at the thresholds of thermal damage: Workshop report by an to projects on critical problems related to radiation technologies ICNIRP task group. Health Phys. 111(3):300–306. in various fields including food, health, water and environment) and (3) energy and electricity (assessing of plans for energy needs International Commission on Radiological Protection (ICRP) including nuclear). Medical applications fall under the first two (Radiation Protection) The International Commission on topics. Radiological Protection (ICRP) is an independent non-gov- In particular, the division of human health, which belongs to ernmental organisation founded as the ‘International x-ray nuclear science, is completely devoted to medical applications. and Radium Protection Committee’ in 1928 by the Second The division is divided into four sections: (1) nuclear medicine; International Congress of Radiology. The aim of the commission (2) applied radiation biology and radiotherapy; (3) dosimetry and is to be the principal body to provide an appropriate international medical radiation physics and (4) nutritional and health-related standard of protection for man without unduly limiting the ben- environmental studies. efit of the practices using ionising radiation. Although originally Hyperlinks: IAEA; http://www .iaea .org established to consider medical exposure to ionising radiation, in International Conference on Radiation Protection 492 International Commission on Radiation Units practice the scope has over the years widened from medical appli- ICRPM 2017 activities: cations to include all exposures to ionising radiation from natural and artificial sources. • Several international organisations and associations The commission is supported by a number of international launched specific actions to support ten proposed organisations and governments. The main activity is to issue rec- actions in the Bonn Call for Action ommendations as an advisory body. The recommendations are • European Commission activities – EUROATOM made available to regulatory agencies at international, regional treaty, BSS, non-binding recommendations and and national levels and they provide guidance on the fundamental communications principles on which radiation protection laws and regulations can • United Nations Scientific Committee on the Effects be based. On the basis of these recommendations, other interna- of Atomic Radiation (UNSCEAR) activities – Global tional organisations and regional/national authorities issue more Medical Exposure Survey detailed regulations and codes of practice. • US Food and Drug Administration (FDA) activities - Due to its historical origin, ICRP is an independent charity introduce safety features into the national and interna- (not-for-profit organisation) registered in the United Kingdom, tional standards for medical devices with a secretariat based in Stockholm, Sweden. It is composed • AFROSAFE RAD activities – radiation safety in Africa of a main commission and five standing committees. The main campaign commission consists of 12 members and a chairman, and the five standing committees cover: (1) radiation effects; (2) doses from ICRPM 2017 introduced the Bonn Call for Action Implementation radiation exposure; (3) radiation protection in medicine; (4) appli- Toolkit – a live online platform with links to different initiatives. cation of commission’s recommendations; and (5) protection of The Bonn Call for Action Implementation Toolkit offers existing the environment. Each of these standing committees is chaired and new implementation resources for improving radiation pro- by a commission member, typically comprises 15–20 members, tection in medicine. and they are collectively served by a small scientific secretariat. Related Articles: Bonn Call for Action ICRP may appoint working parties and task groups to aid in Further Readings: www .i aea .o rg /re sourc es /rp op /re sourc es / the preparation of reports. Working parties are formed by the bo nn -ca ll -fo r -act ion -p latfo rm; standing committees (with the approval of the main commission) ICRPM. 2017. IAEA, Vienna, https://event .do /iaea /a/# /events to develop ideas, and these will sometimes lead to the formation /2009 /f /6371; ICRPM. 2017. Statement, IAEA, Vienna, www .i of task groups. Task groups usually contain specialists from out- aea .o rg /si tes /d efaul t /fil es /do cumen ts /rp op /bo nn -ca ll -fo r -act ion -s side the commission and are assigned responsibility for the prepa- tatem ent .p df; www .iaea .org. ration of draft reports. It remains the responsibility of the main commission to approve final reports for publication. International Commission on Radiation The commission issued its first report in 1928, which was Units and Measurements (ICRU) subsequently numbered publication no. 1 (the recommendations (Radiation Protection) The International Commission on of publication no. 1 were adopted in 1958). Reports have been Radiation Units and Measurements (ICRU) was founded, in 1925, published on general and specialised topics related to radiation by the 1st International Congress of Radiology. Since the begin- I protection and are published as the Annals of the ICRP. ning the principal objective has been the development of recom- Financing of the ICRP comes mainly from voluntary contribu- mendations regarding: (1) quantities and units of ionising radiation tions from international and national bodies; together with royal- and radioactivity; (2) procedures for the measurements and appli- ties from the commission’s publications. Other forms of support cation of these quantities in radiation therapy, diagnostic radiol- come from member institutions. ogy, radiation biology and industrial operations and (3) physical Related Article: International Commission on Radiation Units data needed in the application of the procedures. Therefore the and Measurements (ICRU) ICRU collects and evaluates the latest data and information on Hyperlink: ICRP; http://www .icrp .org the field of radiation protection and dosimetry and prepare rec- ommendations for the most acceptable values and techniques for International Conference on Radiation current use. It

is essential to recognise accepted values for certain Protection in Medicine (ICRPM) parameters in research work and application using ionising radia- (Radiation Protection) The International Conference on tion The first task of the ICRU was that to devise a unit of ionising Radiation Protection in Medicine: Achieving Change in Practice radiation that would make it possible to develop cancer treatment. (11–15 December 2017, Vienna, Austria) is the first official global Afterward many extensive sets of units and quantities have been follow-up of the Bonn Call for Action. developed, as applications became more complex and other fields The 2017 Conference reviewed the actions taken and devel- of application were introduced. opments since the 2012 Bonn conference and acknowledged that The commission consists of 13 members selected for their sci- intensified work in the area of radiation protection in medicine entific merits, regardless of nationalities, and is assisted by some is conducted in response to the Bonn Call for Action, which pro- twenty report committees which work on more ad hoc subjects. vides for a decade long global roadmap on radiation protection in The committees (four to eight members) are appointed to pro- medicine. duce draft documents and may also use external consultants. Two ICRPM 2017 statistics: members of the commission ensure the connection between the committees and the commission. • Organised by the IAEA ICRU is a not-for-profit organisation with a small secretariat. It • Co-sponsored by the WHO and PAHO is supported by a number of international organisations, intergov- • Over 500 participants ernmental bodies, foundations, industries, professional societies, • 97 countries etc. Circa one third of the budget comes from royalties for the • 16 organisations publication of the recommendation. International Electrotechnical Commission (IEC) 493 International Organization for Medical Physics (IOMP) Diagnostic imaging has become increasingly complex with International Federation for Medical and high level of image elaboration, which requires common con- Biological Engineering (IFMBE) cepts, terminology and methodology. The ICRU has enlarged its (General) IFMBE, the International Federation for Medical and programme, ranging from fundamental to practical aspects for Biological Engineering, is primarily a federation of national all types of imaging techniques (including image quality) and has and transnational organisations. These organisations represent published reports on subjects related to modern diagnostic proce- national interests in medical and biological engineering. The dures related to x-ray and nuclear medicine investigations includ- objectives of the IFMBE are scientific, technological, literary and ing also magnetic resonance imaging. educational. Within the field of medical, biological and clinical In order to make real progress in radiotherapy it is essential to engineering IFMBE’s aims are to encourage research and the compare results and methodology. Therefore a common language application of knowledge, and to disseminate information and for reporting doses, techniques, fractionation schedule is crucial. promote collaboration. IFMBE represents the interests of bio- The ICRU has put considerable effort in the definition of appro- medical engineering profession in the UN and WHO as a non- priate guidance levels (for defining tumours, target and planning governmental organisation. volumes) and has given recommendations for the reporting of Membership of all 60 affiliated societies from 55 countries various treatments modalities. constitutes Federation membership which is estimated to 120,000 Careful measurements are required for the protection of work- individuals. IFMBE is affiliated with the International Union for ers. The presence of ionising radiation and radioactive material Physical and Engineering Sciences in Medicine. in the environment influence the protection of the public and the Hyperlinks: www .ifmbe .org environment. Due to diversity in routine and accidental exposures, international measurement standards are required for assessment International Organization for Medical Physics (IOMP) of individual exposures and associated risks. All these aspects (General) The International Organization for Medical Physics have been dealt with extensively. (IOMP) represents about 30,000 medical physicists worldwide, The fundamental aspects of radiation science such as interac- who are members of IOMP’s 86 affiliated national member organ- tion of ionising radiation with matter are studied continuously. isations (as of 2020). The mission of IOMP is to advance medical The related data are necessary in research on mechanisms of physics worldwide by disseminating scientific and technical infor- physical, chemical and biological changes following irradiation, mation, fostering the educational and professional development of and therefore also applications in medicine. medical physics and promoting the highest quality medical ser- Related Article: International Commission on Radiological vices for patients. Protection (ICRP) Medical physics is a branch of applied physics, pursued by Hyperlink: ICRU; http://www .icru .org medical physicists, which uses scientific principles, methods and techniques in practice and research for the prevention, diagnosis International Electrotechnical Commission (IEC) and treatment of human diseases with a specific goal of improv- (Radiation Protection) The International Electrotechnical ing human health and well-being. Commission (IEC) was founded in 1906 with British scientist A medical physicist is a person who is qualified with a univer- Lord Kelvin as its first president. IEC is the leading global organ- sity degree majoring in physics with specialised education and isation that prepares and publishes international standards for all training in the concepts and techniques of applying physics in I electrical, electronic and related technologies. These serve as a medicine and healthcare. basis for national standardisation and as references when drafting The first medical physics society in the world was formed in international tenders and contracts. 1943 in the UK: the Hospital Physicists Association (currently, Through its members, the IEC promotes international coop- the Institute of Physics and Engineering in Medicine, IPEM). The eration on all questions of electrotechnical standardisation and largest medical physics society in the world is in the USA – the related matters, such as the assessment of conformity to standards, American Association of Physicists in Medicine (AAPM, formed in the fields of electricity, electronics and related technologies. in 1958). The IEC charter embraces all electrotechnologies including The International Organization for Medical Physics (IOMP) electronics, magnetics and electromagnetics, electroacoustics, was formed in 1963 in the UK by the societies of Sweden, USA, multimedia, telecommunication and energy production and dis- Canada and the UK. IOMP has the following objectives: tribution, as well as associated general disciplines such as termi- nology and symbols, electromagnetic compatibility, measurement 1. To organise international cooperation in medical phys- and performance, dependability, design and development, safety ics and to promote communication between the various and the environment. branches of medical physics and allied subjects. The commission’s objectives are to 2. To contribute to the advancement of medical physics in all its aspects. • Meet the requirements of the global market efficiently 3. To advise on the formation of national organisations of • Ensure primacy and maximum world-wide use of its medical physics in those countries that lack such organ- standards and conformity assessment schemes isations, and also the possible formation of national • Assess and improve the quality of products and services committees in those countries where there is more than covered by its standards one medical physics organisation. • Establish the conditions for the interoperability of com- plex systems In 1980, IOMP, together with the International Federation of Medical • Increase the efficiency of industrial processes and Biological Engineering (IFMBE), formed the International • Contribute to the improvement of human health and Union of Physical and Engineering Sciences in Medicine (IUPESM) safety as an umbrella body. The IUPESM became a member of the • Contribute to the protection of the environment International Council of Scientific Unions (ICSU) in 1999. International Radiation Protection Association (IRPA) 494 International Radiation Protection Association (IRPA) In 2005, IOMP formed a liaison with the International Union • MEFOMP (Middle East) – including societies from 12 of Pure and Applied Physics (IUPAP) to strengthen the collab- countries with about 900 members oration of medical physicists with other physicists with similar • ALFIM (Latin America) – including societies from 11 interests. Within the IUPAP organisation, IOMP is recognised as countries with about 1,200 members an Affiliated Commission and is referred to as the International • FAMPO (Africa) – including societies from nine coun- Commission on Medical Physics (IComMP). Within IOMP, tries with about 700 members the International Commission on Medical Physics (IComMP) Committee facilitates the collaboration of medical physicists with There is no regional organisation for North America, where physicists who have similar academic interests in research and the medical physics societies of the USA (AAPM) and Canada education. (COMP) have many joint activities. Both have about 10,000 IOMP is a non-governmental organisation (NGO) to the IAEA members. (International Atomic Energy Agency) and the WHO (World IOMP and its sister organisation IFMBE (International Health Organization). IOMP has Memoranda of Understanding Federation for Medical and Biological Engineering) form a union with various related organisations. – IUPESM (the International Union for Physical and Engineering IOMP undertakes a broad range of activities – scientific, edu- Sciences in Medicine). The IOMP has two major meetings – the cation and training and dealing with professional matters. There triennial World Congresses of Medical Physics and Biomedical is a particular emphasis on supporting the development of medi- Engineering, organised through IUPESM (jointly with IFMBE), cal physics in developing countries. The organisation includes and also the International Conferences of Medical Physics, six committees: Scientific Committee; Education and Training held between World Congresses and more regional in nature. Committee; Professional Relations Committee; Awards and Individual sessions or seminars are also organised at other inter- Honours Committee; and Publications Committee; as well as two national meetings. Boards: Medical Physics World Board and Regional Coordination Related Articles: EFOMP, AFOMP, SEAFOMP, MEFOMP, Board. The members of all Committees and Boards are around ALFIM, FAMPO, IFMBE, IUPESM 100 (eminent medical physicists from various countries). The Further Readings: Niroomand-Rad, A., C. Orton, P. Smith elected IOMP Officers and Chairs of Committees and Boards and S. Tabakov. 2013. A history of the international organisation form the IOMP Executive Committee. for medical physics – 50 years anniversary – part I. J. Med. Phys. In 2017, IOMP formed a legal body to represent it – IOMP Int. 1(2):113–116; Niroomand-Rad, A., C. Orton, P. Smith and Company with identical objectives to the organisation. The IOMP S. Tabakov. 2014. A history of the international organisation for Statutes and Bylaws to govern the way the organisation operates. medical physics – 50 years anniversary – part II. J. Med. Phys. The Company members are the members of the IOMP Executive Int. 2(1):7–17; Tabakov. S. 2018. IOMP company report. Med. Committee. Phys. World 34(1):9–10. IOMP has two main regular publications – the newsletter, Hyperlink: www .iomp .org Medical Physics World (launched in 1984 and available free from www .iomp .org) and the journal, Medical Physics International International Radiation Protection Association (IRPA) (launched in 2013 and available free from www .mpijournal .org). (Radiation) The real beginning of the International Radiation I Other journals related to medical physics are also recognised as Protection Association (IRPA) starts as an initiative of the official publications of IOMP such as: Physics in Medicine and American Health Physics Society to form a committee to deal Biology; Physiological Measurement; Medical Physics; Journal with radiation protection problems at international level. After of Applied Clinical Medical Physics; Physica Medica – European several years of work and a much growing interest in this field, Journal of Medical Physics. it was decided to create an international health physics society IOMP has three categories of membership: to associate national societies. The first pro tempore general assembly was held in Paris in 1964, with the participation of 45 1. Individual members: These are the members of national delegates from 15 national health physics or radiation protection organisations associated with adhering national bodies. societies. At the Paris meeting a constitution was adopted and the 2. Adhering bodies: National medical physics organisa- primary objectives were decided. tions or societies. The primary objective of IRPA is to provide a medium 3. Affiliated organisations: Regional organisations and whereby radiation protection professionals from all countries may corporate members. communicate with each other and in this way support the devel- opment and advance of radiation protection. Other knowledge relevant fields and scientific aspects related to radiation protec- The National Medical Physics Societies/Associations members of tion such as medicine, engineering, technology, law, etc. are also IOMP are listed in the articles for IOMP Regional Organisations included. This is in order to provide for a better protection from (see Related Articles). In order to coordinate more effectively,

the the hazardous effects of ionising radiation and at the same time professional activities globally, IOMP has formed six regional facilitate and optimise its beneficial uses. organisations (federations), related to specific continents/regions Other objectives of IRPA include (1) encourage the forma- – as per 2019: tion of radiation protection societies with the purpose to improve international co-operation, (2) support international meetings, • EFOMP (Europe) – including societies from 34 coun- (3) encourage international publications, (4) encourage research tries with about 10,000 members and education in radiation protection and related topics and (5) • AFOMP (Asia and Oceania) – including societies from encourage the establishment and continuous review of universally 18 countries with about 6,500 members accepted radiation protection standards and recommendations. • SEAFOMP (South-East Asia) – including societies The main organisational structure of IRPA consists of the from six countries with about 1,400 members associated societies, the general assembly and the executive International Science Council (ISC) 495 Interruption of treatment council, with a secretariat and a treasury. In addition there are IUPESM is sponsoring and coordinating the triennial ‘World five commissions and committee dealing with various topics. The Congress for Medical Physics and Biomedical Engineering’ executive functions of IRPA are performed by the officers upon (www .iupesm .org). approval of the executive council. An international conference is Related Articles: IOMP, IFMBE, IUPESM held every 4 years. The conference represents a major forum for Further Readings: Smith, P. and F. Nuesslin. 2013. Benefits discussion in the field of radiation protection. to medical physics from the recent inclusion of medical physi- Hyperlink: IRPA; www .IRPA .org cists in the international classification of standard occupations (ICSO-08). J. Med. Phys. Int. 1(1):11 (available free from: www International Science Council (ISC) .m pijou rnal. org /p df /20 13 -01 /MPI- 2013- 01 -p0 10 .pd f); Goh, J., S. (General) The International Science Council (ISC) is a non- D. Tabakov. 2020. 40 years IUPESM. Health Technol. 10:1331– governmental organisation with a unique global membership that 1336, https :/ /do i .org /10 .1 007 /s 12553 -020 - 00493 -8 brings together 40 international scientific unions and associations Hyperlink: www .iupesm .org and over 140 national and regional scientific organisations includ- ing academies and research councils. Interpolation ICS was created in 2018 following the merger of the (Nuclear Medicine) Interpolation refers to the method of calculating International Council of Scientific Unions (ICSU) and the a value in a point in-between two known points in a discrete data set. International Social Science Council (ISSC). The most common form of interpolation is linear interpolation where ICSU was an international non-governmental organisation all the values are estimated from a straight line between the two known devoted to international cooperation in the advancement of sci- points. Interpolation can also be seen as a specific case of curve fitting ence. Its members were national scientific bodies and interna- where the curve must pass through each known data point. tional scientific unions. ICSU was one of the oldest non-governmental organisations in Interruption of treatment the world. The International Union for Physical and Engineering (Radiotherapy) Generally, radiotherapy treatment is delivered Sciences in Medicine (IUPESM) was a member of ICSU from once a day, 5 days a week for up to 8 weeks. Many of the regi- 1999, hence it is now a member of ICS – an organisation that mens currently in use have developed as a result of expediency unites scientific bodies at various levels across the social and rather than from radiobiological principles, evolving to accom- natural sciences. modate the standard 5 day working week. They are considered to Hyperlink: www.council.science compensate empirically for tumour repopulation during the non- treatment weekend breaks but the addition of further interruptions International Union for Physical and Engineering to the planned treatment schedule may result in prolongation of Sciences in Medicine (IUPESM) the overall treatment time. There is extensive evidence that the (General) The International Union for Physical and Engineering prolongation of treatment increases the risk of local recurrence in Sciences in Medicine (IUPESM) was formed in 1980 as a union a wide range of fast-growing tumours and it should therefore be between the International Federation for Medical and Biological avoided. However, there are occasions where unforeseen interrup- Engineering (IFMBE) and the International Organization for tions occur such as machine breakdown or illness of the patient. Medical Physics (IOMP). Through its both constituent organisa- There have been few studies on the effect of treatment prolonga- I tions, IUPESM has about 150,000 members (as per 2020). tion on patients with slow-growing tumours but it is expected that The principal objective of IUPESM is to contribute to the any interruption to a radiotherapy schedule may affect outcome. advancement of physical and engineering sciences in medicine In the United Kingdom, the Royal College of Radiologists has for the benefit and well-being of humanity. Its objectives include addressed the issue of treatment interruptions and published guid- organising international cooperation and promoting communi- ance for clinical practice. They recommend that all radiotherapy cation among those engaged in healthcare science and technol- departments establish robust systems of service planning to cope ogy, coordinating activities of mutual interest to engineering and with predictable and unpredictable interruptions to normal treat- physical science within the healthcare field, including interna- ment including working across bank holidays and the provision tional and regional scientific conferences, seminars, working of adequate resources in terms of machines, staff and training. groups, regional support programmes and scientific and tech- Where an interruption does occur, they recommend one of the nical publications and representing the professional interests following courses of action is taken, listed in order of preference, and views of engineers and physical scientists in the healthcare with priority given to patients with rapidly growing tumours community. being treated with radical intent: IUPESM represents the combined efforts of all medical physi- cists and biomedical engineers (members of IOMP and IFMBE) 1. If due to machine breakdown, transfer of patient to a working in the physical and engineering science of medicine. matched linear accelerator on the day of interruption. The main publication of IUPESM is Health and Technology. 2. Missed weekday treatment fraction to be delivered at IUPESM has been a full member of the International Council the weekend. of Science Union (ICSU) since 1999. During the 2000s, IUPESM 3. Patient treated twice daily with a minimum of 6 h organised a successful campaign that resulted in including the between fractions, preferably towards the end of the occupation of ‘medical physicist’ and the occupation of ‘bio- week to allow repair of sub-lethal normal tissue dam- medical engineer’ for the first time in the International Standard age over the ensuing weekend. Twice daily treatment Classification of Occupations (ISCO-08), published by the is not recommended when fraction size is significantly International Labour Organisation (ILO). Medical physicists are greater than 2.2 Gy. listed under Unit Group 2111, while biomedical engineers are 4. Use of biologically equivalent dose (BED) calcula- listed under Unit Group 2149. This was the most important inter- tions to derive an alternative schedule with a modified national official recognition of both professions. number of treatment fractions in order to complete the Intersource shielding 496 Interstitial brachytherapy radiotherapy course in the original planned overall time. 5. The addition of extra treatment fractions where it is not possible to maintain the original planned overall time even with compensation. The compensation options available will depend on when dur- ing a course of treatment an interruption occurs. The later this is, the less likely it is that the original overall treatment time can be maintained. It should be noted that in cases 3–5, the BED to normal tissues may be higher than that for the original treatment prescription. For cases 4 and 5, a compromise will usually be required between reducing the BED to the tumour and increas- ing the BED to normal tissues. Compensation requiring the use of radiobiological-based calculations should only be adopted when other methods of compensation cannot be applied since assumptions need to be made for parameter values and should only be carried out by appropriately trained physicists or clini- FIGURE I.33 Seed implant, 125I-seeds, intersource effects. cians. A discussion of these methods can be found in the paper by Dale et al. along with a number of practical examples. These examples are reproduced in Appendix B of the Royal College of Interstitial Technique: Radiologists’ report. Related Articles: Biological effective dose (BED), • Applicators/sources are placed inside the target volume Fractionation, Repair, Repopulation, The 5 R’s of radiobiology, (you make the ‘cavities’ yourself) Tumour control probability. • Interstitium (Latin), meaning ‘a space between’, inter Further Readings: Dale, R. G. et al. 2002. Practical meth- (Latin), meaning between ods for compensating for missed treatment days in radiotherapy, with particular reference to head and neck schedules. Clin. Oncol. See Interstitial brachytherapy 14:382–393; The Royal College of Radiologists. 2008. The Timely Related Articles: Brachytherapy, Interstitial brachytherapy, Delivery of Radical Radiotherapy: Standards and Guidelines for Intracavitary brachytherapy the Management of Unscheduled Treatment Interruptions, 3rd edn., Royal College of Radiologists, London, UK. Interstitial brachytherapy (Radiotherapy, Brachytherapy) Placement of Sources in Intersource shielding Brachytherapy: In brachytherapy, applicators and sources are (Radiotherapy, Brachytherapy) placed inside or close to the target volume using the following I Intersource Shielding – Intersource Effect: The source mod- techniques: els used in treatment planning systems describe the dose distribu- tion around the source in a homogeneous ‘infinite’ medium. The 1. Intracavitary technique dose distribution calculations in the treatment planning systems a. Applicators/sources are placed in existing cavities, include no corrections for heterogeneities of any kind (2009). inside or close to the target volume Consequently, when many sources are implanted, as in seed b. The term is derived from intra (Latin), meaning implantations for prostate cancer, the seeds will shield each other. ‘within, inside’ Thus, the total dose is overestimated when calculated as a super- c. This technique is traditionally used for gynaeco- position of the dose distributions from the individual sources. An logic tumours estimation of the reduction of the dose to the prostate shows that i. Applicators/sources of different design are a reduction of several per cent is possible. placed in the vagina and/or the uterine cavity It is to be noted, that all treatment planning systems today ii. There are a large number of applicator types calculate dose this way. Thus, when dose calculation algorithms designed for brachytherapy of cancer of the developments include heterogeneity corrections, in particular cervix intersource effects, this will give a more accurate calculation of iii. For tumours in the vaginal mucosa, vaginal given doses. In addition it will be possible to translate existing cylinders are used dose levels used for specification of prostate seed implants to new d. A subcategory of this technique is the intralumi- and more accurate values (Figure I.33). nal technique, lumen (Latin), meaning ‘passage For an afterloading system with one stepping source, there are within a tubular organ’, for example, intralumi- no intersource effects. nal brachytherapy of the bronchus – endobron- Related Articles: Treatment planning systems – brachyther- chial brachytherapy, where applicator/sources apy, Source models are placed in the bronchus (endon [Greek], mean- ing ‘within’) Interstitial e. As placement of sources is restricted to existing (Radiotherapy, Brachytherapy) cavities, dose gradients are larger in intracavi- Placement of Sources in Brachytherapy: In brachytherapy, tary brachytherapy than in interstitial techniques applicators and sources are placed inside or close to the target (see point 2), where sources are distributed ‘more volume using different techniques. freely’ Interstitial implant 497 Interstitial therapy f. The intracavitary technique is applicable only for temporary implants EXAMPLE 2 2. Interstitial technique a. Applicators/sources are placed inside the target Prostate cancer: Interstitial interactive temporary implant volume (you make the ‘cavities’ yourself) using high dose rate brachytherapy b. Interstitium (Latin), meaning ‘a space between’, The procedure is ultrasound guided, using a template inter (Latin), meaning between for needle positioning, and a dedicated treatment plan- c. Applicators used are ning system to plan and follow the implantation (see also i. Needles, guided by a template; for prostate cancer Temporary implant). (see following section), mammary cancer, etc. These needles are closed, to ensure integrity of the high ii. Catheters; e.g. for head and neck tumours intensity 192Ir source of the remote controlled after loading d. This technique is traditionally used for other

sites unit used for the treatment. than the gynaecological region A preliminary dose distribution with ideal needle posi- e. It is applicable for both permanent and temporary tions is the starting point. After insertion of all the needles, implants the needle positions are adjusted in the treatment planning 3. Surface applications system, a final plan is made (see Figure I.35) and approved, a. In principle a border case of intracavitary technique the patient is treated, and the needles are taken out. Brachytherapy Applications – Interstitial Techniques: EXAMPLE 1 Related Articles: Brachytherapy, Intracavitary brachytherapy, Permanent implant, Temporary implant, Treatment planning sys- Prostate cancer: Interstitial interactive permanent implant tems – brachytherapy of 125I seeds, very low dose rate brachytherapy The procedure is ultrasound guided, using a template for needle positioning, and a dedicated treatment plan- Interstitial implant ning system to plan and follow the implantation (see also (Radiotherapy, Brachytherapy) Placement of Sources in Permanent implant). Brachytherapy: See Interstitial brachytherapy Figure I.34 shows the dose distribution as planned with Related Articles: Brachytherapy, Interstitial brachytherapy, ideal needle positions. Intracavitary brachytherapy, Permanent implant, Temporary During the implantation, needle (and seed) positions in implant the treatment planning system are adjusted to the real nee- dle (and seed) positions interactively. The resulting dose Interstitial therapy distribution, which is continuously updated, is followed (Radiotherapy, Brachytherapy) Placement of Sources in closely during the implant procedure, and extra seeds can Brachytherapy: See Interstitial brachytherapy be added as desired. I Related Articles: Brachytherapy, Interstitial brachytherapy, Intracavitary brachytherapy FIGURE I.34 Dose distribution as planned with ideal needle positions, using a dedicated treatment planning system (VariSeed, Varian). Interventional MRI 498 Intracavitary brachytherapy ideal access. Furthermore, many medical devices designed for conventional interventional use are poorly visualised on MRI, or else result in image degradation on account of their magnetic properties and/or electrical conductivity. Many devices, i.e. tools or implants, also present safety con- cerns in the MRI environment, so care has to be taken to use only MR-safe devices. Special versions of some interventional devices have been developed for MRI use, but as yet there is little com- mercial motivation for this. There are also wider safety concerns relating to the involvement of members of staff, such as surgeons and nurses, who may not be familiar with MRI safety issues, so rigorous safety procedures and training are prerequisite. Interventional MRI has required, and has fuelled, the develop- ment of real-time imaging techniques and rapid reconstruction algorithms, in order to provide the multi-frame per second tempo- ral resolution needed for interventional guidance. Related Article: Real-time imaging Further Readings: Blanco, R. T. et al. 2005. Interventional and intraoperative MRI at low field scanner – A review. Eur. J. Radiol. 56:130–142; Kettenbach, J. et al. 2000. Interventional and FIGURE I.35 The final dose distribution after needle position adjust- intraoperative magnetic resonance imaging. Annu. Rev. Biomed. ment (BrachyVision, Varian). Eng. 2:661–690. Intracavitary (Radiotherapy, Brachytherapy) Placement of Sources in Brachytherapy: In brachytherapy, applicators and sources are placed inside or close to the target volume using different techniques. Intracavitary Technique: Applicators/sources are placed in existing cavities, inside or close to the target volume Intra (Latin), meaning within, inside See also Intracavitary brachytherapy Related Articles: Brachytherapy, Intracavitary brachytherapy, Interstitial brachytherapy Intracavitary I (Ultrasound) Intracavitary (also referred to as intracavity) trans- ducers are designed to image from within a body cavity. These allow placement of the transducer elements near to the target tis- sue and enable improved image quality. Commonly used transducers are FIGURE I.36 Paediatric cardiologist performing MR-guided cardiac 1. Endorectal transducers, used to image the prostate catheterisation using a 1.5 T scanner. gland 2. Endovaginal transducers, used in gynaecology exami- nations and early pregnancy Interventional MRI 3. Transoesophageal transducers for cardiac imaging (Magnetic Resonance) Interventional MRI (iMRI) refers to the performance of invasive clinical procedures under MRI guidance. Various designs are used, usually with tightly curved or phased This is currently a small field with a diverse range of applications, arrays. Arrays can be aligned along the axis of the transducer and/ including minor interventional radiology procedures (such as or transversely across. The use of intracavitary transducers allows breast biopsies), minimally invasive procedures (such as tumour high frequencies to be used. The available aperture is limited by ablation and interventional cardiology) (Figure I.36) and even the physical restrictions necessary to allow penetration into the open neurosurgery. body cavities (Figure I.37). Interventional MRI is developing as an alternative to con- ventional x-ray-guided interventional procedures. Advantages Intracavitary brachytherapy include lack of ionising radiation exposure to patients and staff, (Radiotherapy, Brachytherapy) Placement of Sources in better soft tissue visualisation and contrast, multi-planar imaging Brachytherapy: In brachytherapy, applicators and sources are capability, and the ability to collect images depicting physiologi- placed inside or close to the target volume using the following cal or mechanical information. techniques (Figures I.38 through I.42): There are also significant disadvantages. Conventional MRI systems allow little access to the patient, and although open sys- 1. Intracavitary technique tems are available these are usually low field systems with sig- a. Applicators/sources are placed in existing cavities, nificantly compromised image quality, and most still do not offer inside or close to the target volume Intracavitary brachytherapy 499 Intracavitary brachytherapy FIGURE I.37 Intracavitary transducer designed for endovaginal imag- ing. The tightly curved array (left) images forwards. The handle (right) allows the probe to be manoeuvred. FIGURE I.40 Vaginal cylinder, Lund, high dose rate (Varian). FIGURE I.38 Cervical cancer – Lund old Radium applicator, coupled box and probe. I FIGURE I.41 Vaginal cylinder with intrauterine probe, Lund, high dose rate (Varian). iii. For tumours in the vaginal mucosa, vaginal cylinders are used d. Subcategory; intraluminal technique, lumen (Latin), meaning ‘passage within a tubular organ’ Example: intraluminal brachytherapy of the bron- chus – endobronchial brachytherapy, where appli- cator/sources are placed in the bronchus (endon FIGURE I.39 Cervical cancer – Lund newer ring applicator, for high [Greek], meaning ‘within’) dose rate treatments, remote controlled afterloading (Varian). e. As placement of sources is restricted to existing cavities, dose gradients are larger in intracavitary brachytherapy than in interstitial techniques, where b. The term is derived from intra (Latin), meaning sources are distributed ‘more freely’ ‘within, inside’ f. Applicable only for temporary implants c. Traditionally used for gynaecologic tumours 2. Interstitial technique i. Applicators/sources of different design are a. Applicators/sources are placed inside the target placed in the vagina and/or the uterine cavity volume (you make the ‘cavities’ yourself) ii. There are a large number of applicator types b. Interstitium (Latin), meaning ‘a space between’, designed for brachytherapy of cancer of the inter (Latin), meaning ‘between’ cervix c. Applicators used are Intracavitary therapy 500 Intraoperative radiation therapy (IORT) Gy/tin (*100 rad/tin) 0.5 1.25 1 2.5 1.5 3.75 2 2.5 5 1.25 3 15 3.5 4 7.5 5 6 A΄ 30 10 20 1.25 2.5 2.5 543 2 3.75 Dos/at pocket A 1.25 >1.8 Gy/tin 1 0.5 U = 22 66g (a) VI = 3 0 68g (c) 1.25 1.25 2.5 1.25 2.5 3.75 5 3.75 5 1.25 7.5 7.5 15 10 10 20 15 20 80 1.25 1.25 2.5 2.5 1.25 1.25 (b) (d) FIGURE I.42 Example of dose distribution for the two Lund cervix applicators. (a) Dose rate distribution for the coupled radium applicators, box and I probe. The dose rate at point A (‘2 cm up and 2 cm out’), is 1.8 Gy/h. (b), (c) and (d) Dose distributions for three orthogonal sections through the ring applicator. Ring diameter with spacer 4.1 cm, intrauterine probe length 4 cm. Dose specification; 5 Gy, 2 mm outside the spacer ventrally and dorsally. Dose at point A 5 Gy, dose to organs at risk (rectum and bladder), nominally 5 mm outside the spacer; 77% of specification dose. i. Needles, guided by a template; for prostate and Intracavitary Technique: mammary cancer, etc. ii. Catheters, e.g. for head and neck tumours • Applicators/sources are placed in existing cavities, d. Traditionally used for other sites than the gynaeco- inside or close to the target volume logical region • Intra (Latin), meaning ‘within, inside’ e. Applicable for both permanent and temporary • Subcategory; intraluminal technique, lumen (Latin), implants meaning ‘passage within a tubular organ’. For exam- 3. Surface applications ple intraluminal brachytherapy of the bronchus, also a. In principle a border case of intracavitary technique denoted ‘endobronchial’ brachytherapy (endon [Greek], meaning ‘within’), where applicator/sources are placed Related Articles: Brachytherapy, Interstitial brachytherapy, in the bronchus Temporary implant Isocentric films, high dose rate intraluminal brachytherapy of a Intracavitary therapy bronchial tumour, target length 5 cm, 9 stop positions, step size (Radiotherapy, Brachytherapy) Placement of Sources in 0.5 cm, dose specified 10 mm from the stop positions (Figures I.43 Brachytherapy: See Intracavitary brachytherapy through I.45). Related Articles: Brachytherapy, Intracavitary brachytherapy, See Intracavitary brachytherapy Interstitial brachytherapy Related Articles: Brachytherapy, Intracavitary brachytherapy, Interstitial brachytherapy Intraluminary brachytherapy (Radiotherapy, Brachytherapy) Placement of Sources in Intraoperative radiation therapy (IORT) Brachytherapy: In brachytherapy, applicators and sources are (Radiotherapy) Intraoperative radiation therapy (IORT) is the placed inside or close to the target volume using different techniques. delivery of a single dose of radiation directly to the tumour bed Intraoral radiography 501 Intraoral radiography during surgery. It can be used as the only radiation treatment that a patient will receive, or to provide a precise boost to the tumour bed in addition to a course of external beam radiotherapy. Historically, IORT was cumbersome, requiring transportation of the patient from an operating theatre to a radiation unit during surgery, or custom-built, shielded operating rooms. Now a variety of commercial, mobile IORT systems are available, either generat- ing megavoltage electrons or kilovoltage photons. Brachytherapy can also be used as a form of IORT. As IORT is delivered at the time of surgery, the tissue to be treated can be clearly visualised, reducing the risk of a geographi- cal miss. IORT doses can be relatively well-targeted, limiting side effects. The dose is eliminated to tissue which usually lies in front of the target and can be minimised to the tissue behind the target, via the application of electrons or photons with appropriate ener- gies. The timing of IORT means that the radiation is delivered before tumour cells have a chance to proliferate post-surgery. Logistically, IORT eliminates the risk of patients not complet- FIGURE I.43 AP film. ing a long course of external beam radiotherapy and can reduce treatment costs compared with a full course of external beam radiotherapy. However, IORT also presents significant technical challenges. Intra-operatively, the tumour bed and appropriate irradiation margins can prove difficult to define. Related Articles: External beam therapy, brachytherapy, Electron ranges, Orthovoltage Intraoral radiography (Diagnostic Radiology) Dentists routinely use radiographs for diagnosis, treatment planning and treatment development. Intraoral radiography is one type of dental x-ray investigation. For this acquisition technique: • The source is a dental x-ray tube • The image detector is inside the mouth Dental x-ray equipment is relatively simple – a small x-ray I tube (often with the stationary anode) and small controllable x-ray generator (often in the same tube housing with extended control panel). Exposure time is usually the only selectable parameter. A fixed kV value is commonly used (60 ÷ 70 kV), FIGURE I.44 Lateral film. though rarely two kV values are available. A spacing cylinder is mounted on the tube housing to obtain an x-ray focus to skin distance at least equal to 200 mm (see Figure I.46). The cylinder includes a fixed lead diaphragm. 0.5 1 0.5 2.5 5 7 15 10 0.5 30 1 60 1 0.5 FIGURE I.45 Dose distribution. FIGURE I.46 Dental x-ray tube. Intravascular irradiation 502 Intravascular ultrasound (IVUS) Nowadays, digital detectors have completely replaced ana- logue image receptors (conventional radiographic film). Different technologies are currently available: imaging sensors based on charge-coupled devices (CCD), photostimulable storage phosphor (PSP) image plates, active pixel sensor (APS) and complementary metal-oxide semiconductor (CMOS). Different intraoral radiographic views are commonly used. Bitewing radiograph depicts the crowns of the teeth, allowing for the detection of interproximal caries. Furthermore, the maxil- lary and mandibular alveolar crests are imaged, giving an assess- ment of periodontal status (Figure I.47). Periapical radiograph is a lateral projection displaying both the crown and root of the tooth and the surrounding bone. Two types of exposure technique are used for intraoral periapi- cal radiograph: FIGURE I.49 Occlusal radiography. • Paralleling technique: The long axis of the tooth and the sensor are both

kept on parallel planes and the central placed inside or close to the target volume using the intracavitary ray of the x-ray beam is perpendicular to the tooth being and interstitial techniques. imaged. Intracavitary Technique: • Bisecting angle technique: The x-ray beam is directed perpendicular to the bisected angle formed by the long • Applicators/sources are placed in existing cavities, axis of the tooth (teeth) to be imaged and by the detector inside or close to the target volume (Figure I.48). • Intra (Latin), meaning ‘within, inside’ • Subcategory; intraluminal technique, lumen (Latin), Occlusal radiography is a view taken with the film positioned in meaning ‘passage within a tubular organ’ the occlusal plane (Figure I.49). Example: intraluminal brachytherapy of the bronchus, also Intravascular irradiation denoted ‘endobronchial’ brachytherapy (endon [Greek], meaning (Radiotherapy, Brachytherapy) Placement of Sources in ‘within’), where applicator/sources are placed in the bronchus. Brachytherapy: In brachytherapy, applicators and sources are An intraluminal technique is also used in intravascular irra- diation, also denoted intravascular brachytherapy or endovascular brachytherapy. Sources are placed inside the lumen of an artery, and the intention of the treatment is to prevent vascular restenosis (i.e. to prevent the vessel from ‘becoming narrow again’). I Both gamma-emitting sources, such as the traditional iridium- 192 source, and beta-emitting sources, such as strontium-90 and phosphorus-32, have been used. One specific physics challenge in intravascular brachytherapy is the dosimetry very close to the sources (distance <2 mm); a region of comparatively smaller interest in ‘traditional’ brachy- therapy dosimetry, where generally distances of the order of 5–10 mm are considered. Some of the earlier randomised trials in intravascular brachy- therapy were performed in the 1990s; covering restenotic lesions in the superficial femoral artery and stenting versus stenting plus brachytherapy of coronary arteries. The results were encouraging FIGURE I.47 Bitewing radiograph. and the use especially of intracoronary brachytherapy to control coronary restenosis grew rapidly. But, drug coated stents became available in the early 2000s. These coated stents were easy to use and gave very good treatment results. As a consequence, the use of brachytherapy in the treatment of cardiac restenosis disap- peared quickly in favour of the treatment where no radiation was used. (For a detailed description of intravascular brachytherapy, stenting, vascular restenosis, etc., the reader is referred to existing literature.) Related Articles: Brachytherapy, Intracavitary brachytherapy, Interstitial brachytherapy Intravascular ultrasound (IVUS) (Ultrasound) There are several reasons to visualise arteries and veins from within. The close proximity to the walls allows the FIGURE I.48 Periapical radiograph. use of high frequency with consequent high resolution, allowing Intravoxel incoherent motion (IVIM) 503 Intrinsic efficiency detailed examination of the vessel wall. This includes informa- Thermal noise may be minimised by cooling the amplifier, tion concerning the extent and acoustic properties of plaques and minimising resistances by keeping all impedances low, using the properties of the vessel walls and histological layers since the reactive components where possible (C and L have no significant scanning range is up to 2 cm. This information is not available thermal noise), and limiting bandwidth by appropriate filtering. from angiography. In addition, IVUS allows the assessment of Shot noise: This additional noise is generated in any circuit stenosis with the measurement of luminal area before and after where current is flowing, due to the statistical fluctuation in atherectomy (Hedrick et al., 1995). current as it is made up from many individual charges moving Frequencies of 20–50 MHz are typically used for intravascu- about: lar ultrasound, with a corresponding wavelength of approximately 0.03 mm. In = 2q IB The obvious concern is that IVUS is invasive, but Wui K. Chong (in Lees and Edward, 1996) states that apart from a few where cases of transient coronary artery spasm, there have been no other The noise current In is given as an RMS value complications reported from the use of IVUS and that the tech- I is the mean current flowing nology is simple and safe. B is the bandwidth of the noise Further Readings: Hedrick, W. R., D. L. Hykes and D. E. q is the unit electron charge (1.6 × 10−19) Starchman. 1995. Ultrasound Physics and Instrumentation, 3rd edn., Mosby, London, UK; Lees, W. R. and A. L. Edward. 1996. As the current increases, the proportion of noise to signal Invasive Ultrasound, Martin Dunitz, London, UK. decreases – thus where possible shot noise can be minimised by avoiding the use of low currents. Intravoxel incoherent motion (IVIM) Flicker (I/f) Noise: This is a further source of noise associated (Magnetic Resonance) The terminology intravoxel incoherent with the flow of current through real components. It is significantly motion (IVIM) refers to the effect related to phase dispersion pres- greater at low frequencies, with noise power being inversely pro- ent within voxels subject to random microscopic motion. In con- portional to frequency. This noise becomes dominant over other trast, intravoxel coherent motion (IVCM) is contributed to constant noise sources at frequencies below a few 100 Hz. Flicker noise flow within a voxel and where a phase angle can be determined can be minimised by selecting components of superior quality. and used to estimate the speed of, e.g. flowing blood. The con- Noise Figure (NF): Amplifiers, especially RF amplifiers, are cept of IVIM was introduced by Denis Le Bihan in the late 1980s commonly given a figure of merit reflecting the additional noise in order to characterise random motion in vivo. IVIM includes that the amplifier introduces: effects from both the water self-diffusion and the microcircula- tion of blood in the capillary bed, i.e. perfusion. As detected by æ SNR a diffusion encoding pulse sequence, these two types of motions NFdB = 10 l in ö ogç ÷ are entirely equivalent, with the only difference that the perfusion è SNRout ø motion is faster. The diffusion coefficient resulting from the per- fusion fraction is sometimes denoted the pseudo-diffusion. The In practice this comparison of the signal-to-noise ratios is a gain- effects from the perfusion motion can be identified in the early part independent guide to how well the amplifier performs (NF = 0 dB of the signal versus diffusion-sensitivity decay curve, but due to being perfect). I the speed of the pseudo-diffusion, this part is attenuated already at In practice these sources of noise are small (nV to μV) and diffusion sensitivities on the order of 300 s/mm2. In conventional interference from other electronic sources will be much greater diffusion imaging where the apparent diffusion coefficient (ADC) than intrinsic noise in most applications. is estimated based on measurements performed with diffusion sen- Abbreviation: NF = Noise figure. sitivity zero (b = 0) and b = 1000–1500 s/mm2, the ADC will be Further Reading: Horowitz, P. and W. Hill. 1989. The Art of overestimated due to the perfusion related signal decay. In cerebral Electronics, Cambridge University Press, Cambridge, UK. imaging this overestimation will be the greatest in the grey matter. Related Articles: Apparent diffusion coefficient (ADC), Intrinsic efficiency b-value (Radiation Protection) The intrinsic efficiency ɛint of a radiation detector is defined as a quotient of the number of recorded pulses Intrinsic amplifier noise Nrec to the number of radiation quanta (particles or photons) Nrad (General) All electronic amplifiers introduce some additional incident on detector: electronic noise to the signal being amplified. The common sources of noise are as follows: N ein = rec t Thermal Noise (Johnson–Nyquist Noise): This is caused by Nrad the thermally generated movement of electrons in conductors and The intrinsic efficiency depends on the detector material, its defined mathematically as thickness, and the type and energy of the incident radiation. The absolute detection efficiency ɛabs is expressed as a ratio Vn = 4kTRB of the number of recorded pulses Nrec to the number of radiation quanta emitted Nem by a radiation point source: where The noise voltage Vn is given as RMS mean voltage N R is the resistance of the conductor eab = rec s Nem T is the absolute temperature B is the bandwidth in Hz of the noise For an isotropic source, the relationship between intrinsic and k is Boltzmann’s constant (1.38 × 10−23 m2 kg/s2/K) absolute detection efficiency is therefore: Intrinsic flood field uniformity 504 Inverse square law correction 4p beam. In complex treatments, such as IMRT, this is often achieved e æ int = e ö abs ç ÷ using an automated, optimisation process. It is called inverse plan- è W ø ning in that the planner specifies the desired dose distribution and where Ω is a solid angle (in steradians – sr) subtended by the the optimisation system computes the optimum beam distribution detector from the position of the radiation point source. to deliver a plan that matches the prescribed dose distribution as Related Article: Detection efficiency closely as possible; thus, it is the opposite of normal, interactive Further Reading: Knoll, G. F. 2000. Radiation Detection and forward planning. The desired dose distribution is usually speci- Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. fied in terms of the desired dose to the target (including allowed 116–117. variation from that dose) and dose limits to normal tissues, plus importance factors for the various tissue regions. In some cases Intrinsic flood field uniformity constraints on the dose volume histogram or desired calculated TCP and NTCP values may be specified. The optimisation is often (Nuclear Medicine) Scintillation camera uniformity measured carried out using a standard least squares gradient descent method without a collimator using a point source (providing a uniform or simulated annealing. Upon satisfactory calculation of the irradiation of the detector surface). Two different parameters are required fluence maps of the IMRT beams, the treatment planning of interest; integral and differential uniformity. Integral unifor- system can then generate for each of the beams a set of multi-leaf mity is defined as a measure of the maximum pixel count devia- collimator (MLC) motion sequence files (dynamic delivery), or a tion in the central field of view (CFOV) and useful field of view series sub fields (step and shoot delivery), which can be transferred (UFOV). This is calculated by dividing the difference between to the MLC controller of the treatment machine for delivery of the the maximum and minimum pixel value by their sum and then treatment beams. multiplying by 100: Abbreviation: CT = Computed tomography. Max - Min Related Articles: Treatment planning, Interactive planning, Integraluniformity = ± ´ æ ö 100 ç Dose volume histogram è Max + Min ÷ (I.3) ø The differential uniformity is defined as the maximum deviation Inverse square law between five contiguous pixels: (Nuclear Medicine) In radiation physics the inverse square law refers to the fact that radiation intensity is inversely proportional Max - Min to the square distance, d from the source: Differential uniformity = ± ´ æ ö 100 ç è Max + Min ÷ (I.4) ø 1 I µ Each row and column in the image is treated as separate slices. A d2 set of five contiguous pixels are selected in the slice, the maximum and minimum value is identified and used to calculate the differen- For example, the intensity at 1 m away from the source is four tial uniformity for that particular set. This procedure is continued times higher than the intensity at 2 m. The law plays a central part for each row and column within the specified area (e.g. CFOV/ in radiation protection, e.g. personal must be trained in handling I UFOV). The maximum differential uniformity is then identified. tweezers in order to increase the distance between the source and Related Articles: Central field of view (CFOV), Useful field hands. of view (UFOV) Further Reading: NEMA. 2001. NEMA Standards Inverse square law correction Publication NU 1-2001, National Electrical Manufacturers (Radiotherapy) It is reasonable to assume that the source of x-ray Association, Rosslyn, VA. photons is a point source, and therefore the beam that is produced is divergent in nature. Therefore assuming a square field of side a Inverse Fourier transform (with an area A = a2) is produced at a distance da from the source (General) The Fourier transform is a very common method to and a square field of side b (area B = b2) is produced at a distance express signals or images as periodical sine and cosine functions db, the two can be related as follows: with an amplitude

and a frequency. The definition of the Fourier transform F of a function f is as follows: a d = a b db ¥ Assuming the source produces a total of N x-ray photons with a F(w) = ò f (t)e-iwtdt fluence of ΦA at da, and fluence ΦB at db from the source. Then the -¥ two can be related as follows: The inverse Fourier transform will then be N = FAA = FBB ¥ 1 f (t) = ò F(w)eiwtdw 2p and -¥ FA B b2 d2 = = = b FB A a2 d2 Inverse radiotherapy planning a (Radiotherapy) Radiotherapy treatment planning involves choos- So the fluence is inversely proportional to the square of the dis- ing the beam directions, field shapes and intensities for the tance from the source, and by extension the exposure, air kerma patient’s beam delivery. This is usually done using a CT scan of in air and dose will likewise be inversely proportional to the the patient and a model of dose deposition in the patient by the square of the distance from the source. It is necessary to allow Inversion recovery 505 Inversion recovery for this inverse square law correction when performing calcula- create a fast inversion recovery sequence type, often also denoted tions at different distances from those where the dose calibration turbo inversion recovery. In this sequence type, the series of RF is performed. For example, typically the linac output is quoted pulses during one repetition are 180°–90°–180°–180°–180° – … . at the depth of maximum dose (zmax) for a square field of side Fast IR sequences have opened the possibility to perform the 10 cm measured at the nominal SSD of the unit (typically 100 IR experiment within reasonable acquisition times, since in this cm). Therefore for isocentric treatment calculations the dose must case Tacq will be reduced by the echo train length (ETL) factor be corrected using an inverse square correction factor (unless the as in TSE. linac is actually calibrated at isocentre) as shown here: It can be shown that the IR signal is approximated by (neglect- ing transversal relaxation) 2 æ SSD + z ISLCorr = max ö ç è SSD ÷ ø S ~ PD (1- 2e-TI /T1 - e-TR /T1 IR ) where TI (inversion time) is the time between the inversion Inversion recovery pulse and the 90° pulse. The signal range in this case, assum- (Magnetic Resonance) In a pulse sequence starting with a 90° ing TR≫T1, TR>TI, goes between −Smax ∼ −PD (TI = 0) and excitation pulse, such as the spin echo (SE, 90°–180°) and the Smax ∼ PD (TI → ∞). Hence the dynamic signal range for IR fast spin echo (FSE, 90°–180°–180°–180°– …), T1 contrast is will be about twice the range for SE, allowing for increased T1 obtained by adjusting the sequence repetition time TR so that contrast (Figure I.50c). Another feature of interest with the IR object parts with different T1 values, due to different longitudinal pulse sequence is that the signal as a function of TI is nulled at recovery during TR, obtain different signal values. The signal a TI which depends upon TR and T1, indicating that object parts ranges between 0 (TR = 0) and Smax ∼ PD (TR → ∞) if transver- with specific and known T1 values can be nulled for a properly sal relaxation effects are neglected. This signal range, creating T1 chosen TI. For very long TR values TInull ≈ 0.69 T1. With respect contrast, can be increased by introducing an inversion pulse (i.e. to image contrast, the IR image can be visualised in two differ- a 180° pulse), thereby inverting the direction of Mz, prior to the ent ways: Using phase-sensitive reconstruction the greyscale can 90° pulse used to flip the magnetisation into the transversal plane be adjusted for negative as well as positive signal values, so that (Figure I.50). This experiment, denoted inversion recovery, was negative signal values will be dark, zero signal grey and posi- early shown to give good clinical (morphological) images (1). In tive signal values bright. This type of display will yield a char- order to prevent susceptibility-induced dephasing, a full SE or acteristic grey background outside the object (zero signal) and FSE sequence can be executed after the inversion pulse instead occasional ‘background-like’ parts inside the object for object of only a 90° pulse. The obvious drawback with spin-echo-based parts with nulled signal. Another representation is provided IR sequences is long acquisition times since commonly high by magnitude reconstruction, where absolute signal values are TR values are chosen. A way of overcoming this problem is to shown. In the latter case, the improved dynamic signal range is Mz 180° SE/FSE I (90°.) TI (a) 1 1 0.5 0.5 S 0 S 0 0 500 1000 1500 2000 2500 3000 3500 0 400 900 1400 1900 2400 2900 3400 T –0.5 I/ms TI/ms –0.5 –1 –1 (b) (c) FIGURE I.50 (a) In an inversion recovery experiment, a 180° RF pulse inverts the longitudinal magnetisation. After a waiting time denoted TI, tissues with different T1 (black, dark grey and light grey arrows) have undergone different amounts of longitudinal relaxation. This difference in potential signal is flipped into the transversal plane for readout, frequently by combining the 180° inversion pulse with a fast spin echo sequence. (b) Using phase-sensitive reconstruction, the greyscale can be adjusted for negative as well as positive signal values, so that negative signal values will be dark, zero signal grey and positive signal values bright. (c) Another representation is provided by magnitude reconstruction, where absolute signal values are displayed. Inversion time (TI) 506 I odine-123 maintained and the background outside the object will become TR dark ‘as usual’, although signal crossing points, i.e. equal sig- TI nal for objects with different T1 values, may appear and may obscure contrast. Furthermore, additional T2 weighting obtained by using a long echo time may enhance contrast for properly EPI EPI chosen TI values. 180° 180° Top: In an inversion recovery experiment, a 180° RF pulse Tag Control inverts the longitudinal magnetisation. After a waiting time denoted TI, tissues with different T1 (black, dark grey and light FIGURE I.51 Schematic diagram of a pulsed ASL sequence. grey arrows) have undergone different amounts of longitudinal relaxation. This difference in potential signal is flipped into the transversal plane for readout, frequently by combining the 180° History: Although it is known that iodine was first isolated in inversion pulse with a fast spin echo sequence. 1811 by Bernard Courtois during the manufacture of saltpetre, Related Articles: Echo train length, Fast spin echo, Repetition there was some debate over who should be credited with identify- time (TR), RF pulse, Spin echo, T2-weighted ing it as a new element. In December 1811, within days of each Further Reading: Bydder, G. M. et al. 1982. Clinical NMR other, Sir Humphry Davy and Joseph Louis Gay–Lussac indepen- imaging of the brain, 140 cases. Am. J. Roentgenol. 139:215–236. dently announced their discovery of the element. Isotopes of Iodine: 37 isotopes of iodine exist. Only one, Inversion time (TI) iodine-127, is stable and exists in nature as part of compounds such (Magnetic Resonance) The inversion time (TI) is the time period as iodides. Several unstable isotopes are used in nuclear medicine from an 180° inversion pulse to the excitation pulse. Sequences imaging and radionuclide therapies. These are as follows: employing an inversion pulse are called inversion recovery pulse sequences. Since different tissues have different T1, the inversion Isotope of iodine 131I time affects the image contrast. Very often the inversion time is chosen to null a certain tissue or fluid for signal suppression, e.g. Half-life 8.02 days of fat using STIR or cerebrospinal fluid using FLAIR. Peak decay energy 606 keV beta (positron); 364 keV gamma Related Articles: Isotope of iodine 125 Fluid attenuated inversion recovery I (FLAIR), Inversion recovery, Short tau inversion recovery (STIR) Half-life 60 days Peak decay energy 35 keV gamma Inversion time delay (TI delay) Isotope of iodine 123I (Magnetic Resonance) The inversion time delay (TI), also called Half-life 13.22 h tag delay or post-labelling delay time, is an MR-specific pulse- Peak decay energy 159 keV gamma; 27 keV gamma; some internal sequence parameter related to the arterial spin labelling (ASL) conversion x-rays concept. This parameter is the time between the arterial spin labelling and the start of the echo-planar imaging (EPI) acquisi- I tion of the imaging slice. In pulsed ASL this time must be longer (Medical Applications) than the transit time of labelled arterial blood from the inversion Nuclear Medicine Diagnostic Imaging: Both stable and site to the imaging slice. The optimal upper limit of TI depends on radioactive iodine are readily taken up by cells in the thyroid the specific pulsed ASL technique employed, but TI should gener- gland. Radioactive isotopes of iodine with suitable emission ally not be excessively long, due to the fact that any unnecessary properties and half-life are used as tracers for diagnostic imag- T1 relaxation of the inverted spin population leads to a reduced ing of the thyroid (and occasionally the kidney, although this is contrast-to-noise ratio. largely obsolete). The tracer, usually iodine-123 or iodine-125 in The inversion time delay is denoted TI in the Figure I.51. the form of sodium iodide, is injected intravenously and the thy- Related Article: Arterial spin labelling (ADC) roid gamma emissions are imaged with a gamma camera to assess Further Reading: Campbell, A. M. and C. Beaulieu. 2006. thyroid function. Pulsed arterial spin labeling parameter optimization for an elderly Radionuclide Therapy: Iodine-131 is administered orally to population. J. Magn. Reson. Imaging 23:398–403. patients with thyrotoxicosis or thyroid cancer, and taken up by the thyroid. The emitted beta radiation ablates thyroid tissue to Iodine reduce thyroid activity or remove cancerous cells. Iodine-131 is (General) also used as part of meta-iodo-benzyl-guanidine (MIBG) therapy to ablate certain neuroendocrine tumours. Iodine-125 is used in low dose rate seeds for prostate brachytherapy. Symbol I Related Articles: Thyroid radioiodine uptake measurement, Element category Halogen Iodine-123, Iodine-125, Iodine-131, Brachytherapy sources Mass number A of stable isotope 127 (100%) Atomic number Z 53 Iodine-123 Atomic weight 126.90447 (Nuclear Medicine) Electronic configuration 1s2 2s2 2p1 3s2 3p6 3d10 4s2 4p6 4d10 5s2 5p5 Element: Iodine (halogen) Melting point 386.85 K Isotopes: 55 < N < 88 Boiling point 457.4 K Atomic number (Z): 53 Density near room temperature 4.933 × 103 kg/m3 (4.933 g/cm3) Neutron number (N): 70 Symbol: 123I Iodine-124 507 I odine-125 Production: cyclotron by 121 123 EC Sb(a,2n) I ® 123Te or Radiation: β+ radiation; γ-radiation followed by x-rays and 127 EC I(p,5n)123Xe ® 123I ® 123Te 13,27h electrons 2,08h Energy: hν = 511 keV (200%) Daughter: 123Te Positron range ∼3 mm in tissue Half-life: 13.27 h Biological half-life: 80 days (thyroid), 8 h (other organs) Decay mode: Electron capture (EC) Critical organ: Thyroid (normal uptake 30%–35%) Radiation: γ-radiation followed by x-rays and electrons ALI Energy: γ 159 keV (83%), 27 keV (71%), 529 keV (1%); min (50 mSv): 100 MBq Absorbed dose (iodide): 4.5 mGy/MBq (thyroid) Conversion electrons: 127 keV (14%) Effective dose: 0.2 mSv/MBq (oral); 0.08 mSv/MBq Shielding: HVL = 1 mm lead; electron range 0.2 mm in (inhalation) glass Biological half-life: 80 days (thyroid), 8 h (other organs) Critical organ: Thyroid (normal uptake 30%–35%) 53 2P° ALI 2– 3/2 4.1760 d min (50 mSv): 100 MBq Absorbed dose (iodide): 4.5 mGy/MBq (thyroid) 124 I EC I Effective dose: 0.2 mSv/MBq (oral); 0.08 mSv/MBq 53 (inhalation) Iodine 126.90447 53 2P3° [Kr]4d10 5s2 5p5 Qβ–294 /2 10.4513 QEC3159.6 I 5/2+ 13.27 h Clinical Applications: Iodine-124 is used for PET-imaging in Iodine 123 diagnostic nuclear medicine. It is labelled to the same pharma- 126.90447 I 53 ceuticals as 131I. [Kr]4d10 5s2 5p5 EC The clinical applications are the same as for 123I-MIBG for 10.4513 the detection of neuroblastoma and pheocytoma, 123I as iodide for imaging of thyroid tissue, and 123I-Ioflupane (DaTScan) for pre- Clinical Applications: Iodine-123 is used for imaging in synaptic neuro-imaging. nuclear medicine because of its nearly ideal photon energy for Related Articles: Iodine-123, Iodine-125, Iodine-131 the scintillation camera, i.e. 159 keV (89%). The radionuclide also Further Readings: Chu, S. Y. F., L. P. Ekström and R. B. emits photons with higher energy resulting in a risk of

septum Firestone. 1999. The Lund/LBNL Nuclear Data Search, [http: / / penetration in the collimator. nuc leard ata .n uclea r .lu. se /nu clear data/ toi/]; Firestone, R. B. 1999. Typical clinical applications are SPECT-imaging with Table of Isotopes, 8th edn., Update with CD-ROM. [http://ie .lbl 123I-MIBG for the detection of neuroblastoma and pheocytoma, .gov /toi .html]; ICRU. 1988. Radiation Dose to Patients from 123I as iodide for imaging of thyroid tissue, and 123I-Ioflupane Radiopharmaceuticals, ICRU Publication 53, Washington, DC; (DaTScan®) for presynaptic neuro-imaging. Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in I Related Articles: Iodine-125, Iodine-124, Iodine-131 Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American Further Readings: Chu, S. Y. F., L. P. Ekström and R. B. Pharmacists Association, Washington, DC; MIRD Dose Estimate Firestone. 1999. The Lund/LBNL Nuclear Data Search, [http: / / Report No. 5. 1975. Summary of current radiation dose estimates nuc leard ata .n uclea r .lu. se /nu clear data/ toi/]; Firestone, R. B. 1999. to humans from I-123, I-124, I-125, I-126, I-130, I-131 and I-132 Table of Isotopes, 8th edn., Update with CD-ROM. [http://ie .lbl as Sodium Iodide. .gov /toi .html]; ICRU. 1988. Radiation Dose to Patients from Radiopharmaceuticals, ICRU Publication 53, Washington, DC; Iodine-125 Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in (Radiotherapy, Brachytherapy) Iodine-125 decays by electron Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American capture to an excited state of tellurium-125 followed by decay Pharmacists Association, Washington, DC; MIRD Dose Estimate to the ground state, giving 35.5 keV gamma rays and 27–32 keV Report No. 5. 1975. Summary of current radiation dose estimates x-rays. The half-life of iodine-125 is 59.4 days and the average to humans from I-123, I-124, I-125, I-126, I-130, I-131 and I-132 energy 0.028 MeV. (Iodine-125 is produced by neutron activation as Sodium Iodide. of xenon-124, naturally occurring, resulting in xenon-125, which decays by electron capture to iodine-125, which in its turn decays Iodine-124 by electron capture to tellurium-125.) (Nuclear Medicine) The most common use of iodine-125 sources is for permanent interstitial implants; it is also used for temporary applications Element: iodine (halogen) with eye plaques. The low energy of iodine-125 minimises the Isotopes: 55 < N < 88 radiation protection problems, and interstitial implants are per- Atomic number (Z): 53 formed using both manual and manual afterloading techniques. Neutron number (N): 71 Note that iodine-125 is also used for in vitro nuclear medicine, Symbol: 124I e.g. in biological assays. Production: Cyclotron There are a number of commercially available designs of Daughter: 124Te iodine-125 sources, often denoted seeds. The designs differ in Half-life: 4.176 days the way the radioactive material is distributed within the encap- Decay mode: EC (74.4%); β+ (25.6%) sulation and the materials used, see Figure I.52. An important Iodine-131 508 I odine-131 4.5 mm 4.7 mm Titanium End weld Resin spheres with128I adsorbed to surface Titanium capsule 0.5 mm outside the spacer capsule Au-Cu markers 0.8 mm 0.8 mm Amersham health model 6702 source NASI model MED3631-A/M or MED3633 source 4.5 mm 4.8 mm End weld Silver rod with128I adsorbed to surface Titanium capsule Titanium capsule Radioactive ceramic Gold marker 0.5 mm 0.5 mm Amersham health model 6711 source Bebig model 125.S06 source 5.0 mm 4.5 mm Titanium Titanium Tungsten Carbon coating outer capsule inner capsule marker containing 128I End weld Silver spheres with128I adsorbed to surface Titanium capsule 0.8 mm 0.8 mm Best model 2301 source Imagyn model IS-I2501 source 4.5 mm Titanium Titanium Graphine pellers with Lasser weld 2nd cup capsule Lead marker 128ICl coating both ends I 0.8 mm erag enics model 200 source FIGURE I.52 Examples of commercially available Iodine-125 seeds. property of a seed is its visibility, as modern interstitial implant Further Reading: Rivard, M. J. et al. 2004. Update of the techniques are image-guided, using ultrasound, fluoroscopy, CT AAPM Task Group No.43 Report: A revised AAPM protocol for or MR. Thus, manufacturers use materials that show the seed brachytherapy dose calculation. Med. Phys. 33:633–374. in ultrasound imaging (surface properties), fluoroscopy and CT (x-ray attenuating properties) and MR imaging. The design of a Iodine-131 seed, the choice of material and the distribution of the radioactiv- (Nuclear Medicine) ity all affect the dose distribution around the seed. Thus, even if the nominal seed strength is the same, seeds of different types and Element: iodine (halogen) from different manufacturers will a priori have different dose dis- Isotopes: 55 < N < 88 tributions. Seeds are available with a variety of source strengths, Atomic number (Z): 53 and as loose seeds and strands, see Figure I.53. Neutron number (N): 78 For an update on commercially available seeds and their Symbol: 131I physical properties, accepted by the brachytherapy community Production: Fission product as complying with AAPM recommendations, see the web site of Daughter: 131Xe the Radiological Physics Centre (RPC): http://rpc .mdandersson . Half-life: 8.04 days org /rpc Decay mode: β−-decay Abbreviations: AAPM = American Association of Physicists Radiation: β− radiation, followed by γ-radiation in Medicine. Beta energy released: 970 keV (max)/323 keV (mean) Related Articles: Brachytherapy sources, Interstitial brachy- Gamma energy: 364.5 keV (81.5%) 80.2 keV (2.6%) therapy, Permanent implant Absorption (HVL): 3 mm lead IOMP 509 Ion recombination Nuclear Pharmacy and Nuclear Medicine, 2nd edn., American Pharmacists Association, Washington, DC; MIRD Dose Estimate Report No. 5. 1975. Summary of current radiation dose estimates to humans from I-123, I-124, I-125, I-126, I-130, I-131 and I-132 as Sodium Iodide. IOMP (General) See International Organisation for Medical Physics Ion (Nuclear Medicine) An ion is an atom where the proton-electron ratio differs from one which gives it the property of either posi- tive or negative charge. An ion with more electrons in the shell structure than protons in the nuclei has a negative charge and is referred to as anion. An ion with positive charge is called a cation (pronounced ‘cat-eye-on’). Ion collection (Nuclear Medicine) See Ion pair FIGURE I.53 Stranded seeds, sterile conditions; upper panel – a two seed strand has been cut; lower panel – the strand is inserted in the implant needle, to be followed by a mandrin. Ion pair (Nuclear Medicine) An ion pair refers to an ionised atom (ion) and its corresponding electron, i.e. the electron that has been removed Biological half-life: 80 days (thyroid), 8 h (other organs) from the atom. Ion pairs have a natural tendency to recombine but Critical organ: Thyroid (normal uptake 30%–35%) in some detector systems the recombination is suppressed and the ALImin (50 mSv): 1 MBq ion pairs are used as a detector response. In an ionisation chamber Absorbed dose (iodide): 530 mGy/MBq (thyroid) detector gas molecules are ionised by incident radiation and the Effective dose: 22 mSv/MBq (oral); 8 mSv/MBq (inhalation) resulting ion pairs are collected using an electric field. Further Reading: Knoll, G. F. 2000. Radiation Detection and Clinical Applications: Iodine-131 has been used in nuclear Measurement, 3rd edn., John Wiley & Sons, New York, p. 30. medicine since the late 1930s, and is the standard radionuclide for thyroid treatment, as sodium radioiodine. After oral admin- istration of radioiodine, the absorption from the GI-tract is rapid, Ion recombination 5% min−1, and nearly completed after 1–2 h. The rate of absorp- (Radiotherapy) When ionising radiation interacts with a gas tion may slow down if food is present, and it is influenced by the the positive ions or free electrons which are created have a ran- function of the thyroid. Iodine-131 is also used for, e.g. treatment dom thermal motion and consequently they diffuse away from regions of high intensity having many types of collision. In I of neuroblastoma and pheochromocytoma as 131I-MIBG (meta- iodobenzylguanidine) and occasionally as labelled monoclonal some collisions an electron may be transferred from a neutral antibodies for treatment of specific cancers. gas molecule to a positive ion thereby reversing their role or an electron may be captured by a positive ion returning to a state of charge neutrality. Collisions between positive ions and free 53 2P electrons may result therefore in a recombination in which the 3°/2 electron is captured by the positive ion and returns it to a state of charge neutrality. Alternatively, the positive ion may undergo a I collision with a negative ion and both ions are neutralised. The 7/2+ 8.02070 d frequency of collisions is proportional to the product of the con- Iodine centrations of the two species involved. The recombination rate 126.90447 131 I 53 can be expressed as [Kr]4d10 5s2 5p5 β– 10.4513 dn = -an+n- dt Related Articles: Iodine-123, Iodine-124, Iodine-125 Further Readings: Annals of the ICRP. 1987. Radiation Dose where to Patients from Radiopharmaceuticals, Biokinetic Models and n+ is the number density of positive species Data, ICRP Publication 53, Vol. 18, Pergamon Press, Oxford, n− is the number density of negative species UK; Annals of the ICRP. 1998. Radiation Dose to Patients α is the recombination coefficient from Radiopharmaceuticals, Vol. 28, No.3, Addendum to ICRP Publication 53, ICRP Publication 80, Pergamon Press, Oxford, Recombination is therefore significant only in regions in which UK; Chu, S. Y. F., L. P. Ekström and R. B. Firestone. 1999. The the concentration of ions and electrons is high and this condition Lund/LBNL Nuclear Data Search, [http: / /nuc leard ata .n uclea r surely can be found along the track of an ionising particle where .lu. se /nu clear data/ toi/]; Firestone, R. B. 1999. Table of Isotopes, the ion concentrations are maximum before the ion diffusion. In 8th edn., Update with CD-ROM. [http://ie .lbl .gov /toi .html]; the presence of an external electric field, as in the case of the sen- Kowalsky, R. J. and S. W. Falen. 2004. Radiopharmaceuticals in sitive volume of an ionisation chamber filled with air and exposed Ion therapy 510 Ion therapy to radiations, the drift of the positive and negative charges repre- sented by the ions and electrons constitutes an electric current and I its intensity is influenced by the recombination process. Electrodes I A recombination occurs when the positive and negative ions V formed in the same track meet and recombine. This process is Gas filled called initial or columnar recombination as the tracks tend to be tube V linear in the gas. Initial recombination is a dose rate indepen- dent effect unless the ion density becomes so great that the field FIGURE I.54 Current–voltage characteristics of an ionisation chamber. strength responsible for the ion collection is reduced. The initial recombination is small for the radiation dose rates encountered in the clinical use of ionising radiation. When ions from different tracks encounter each other on their way to the collecting elec- 1.0 trodes and recombine the effect is called general recombination. This effect depends on how many ions are created per unit volume Q΄= Q at saturation and per unit time and therefore it depends on dose rate since a greater density of ions of both signs increases the probability that they will recombine. 0.8 It is possible to distinguish experimentally between the initial and the general recombination by plotting the reciprocal of the observed ionisation current I against a suitable function of the collecting field strength X. For the initial recombination the function is given by 0.6 1 1 k = + i isat X while for general recombination the relation is 0.4 1 1 k¢ = + i i 2 sat X with k and k′ constant. 0.2 The total recombination is determined by the sum of the initial and general recombinations. In Figure I.54 the current voltage characteristics of an ionisa- tion chamber are reported. No net current flows in absence of an I applied voltage. As the voltage increases the electric field begins 0 0 1 2 3 4 5 6 7 to separate the ions without avoiding the recombination effect. Potential (a.u.) The combined effect of recombination and diffusion determine the shape of the rise to the saturation in the current voltage char- acteristic curve. The ion diffusion effect does not permit mea- FIGURE I.55 Variation of the ionisation charge Q′ collected by an ioni- surement of the saturation current and in practice the voltage V is sation chamber versus applied voltage. raised to high values to minimise the losses. One empirical relationship that permits the evaluation of losses due to the diffusion effect when the applied voltage is suf- possible to increase indefinitely the applied

voltage because of ficiently high is given by ion multiplication in the ionisation chamber gas and because of electrical breakdown of insulator used in the ionisation chamber 1 1 d3 construction. It is necessary therefore to determine a recombina- = + k I IS V 2 tion factor to determine Q at saturation. where Ion therapy IS is the saturated ionisation chamber (Radiotherapy) Ion therapy is radiotherapy delivered with ions d is the electrode separation distance in the chamber of light nuclei, often carbon, to treat deep seated tumours. The V is the applied voltage advantage over x-rays is the fact that the ions have a Bragg peak k is a constant characteristic of the specific chamber which means that the dose is delivered mostly over a small range of depth rather than with the classic exponential distribution of The dose deposited in the air inside the sensitive volume of x-rays. Compared with protons, the energy deposition density is an ionisation chamber is proportional to the charge Q produced greater. by radiations. Because of the recombination the charge Q′ col- A magnetic scanning system may be used to direct the beam lected by the biased electrode of the chamber is less than Q and across the tumour in a raster scan, which coupled with energy it depends on the collecting voltage. In Figure I.55 the variation modulation allows intensity-modulated radiotherapy (IMRT) of the ratio between Q′ and Q is reported as a function of the with the ion beam. Energy modulation allows broadening of the collecting voltage. When the applied voltage increases the recom- depth of maximum dose in what is known as a spread out Bragg bination effect decreases. Q′ = Q at the saturation but it is not peak (SOBP). f = Q΄/Q Ionisation 511 I onisation chamber In order for the ions to penetrate sufficiently deep to treat deep seated tumours, they must have energies of 300–400 MeV/nucleon typically. This is often achieved using large particle accelerators such as synchrotrons, meaning that this form of treatment is often delivered at large national facilities. Ion therapy should not be confused with so-called negative ion therapy. Abbreviations: IMRT = Intensity-modulated radio therapy and SOBP = Spread out Bragg peak. Related Articles: Proton therapy, Spread out Bragg peak, Heavy particle beams, Hadron therapy, Charged particle therapy V Ionisation FIGURE I.57 Example of a current–voltage characteristics of the ioni- (Radiation Protection) The ejection of an electron from an atomic sation chamber. shell by the interaction of radiation. There are a variety of interac- tions that can result in ionisation. For example, x- or gamma-rays interact with matter via a number of different processes, of which two result in ionisation; Compton scattering and the photo-elec- tric effect. Ionization Ionisation presents a biological risk since the absorption of R V chamber R energy both from the incident radiation, and the kinetic energy of the ejected electron within cellular molecules can lead to malfor- mation (as a result of single ionisations of the DNA molecule) or deterministic effects (as a result of gross cell killing from multiple – + ionisations within each cell). V Related Articles: Interaction, Photoelectric effect, Compton R ~ Exposure rate scatter FIGURE I.58 Example of the ion current measurement with the use of Ionisation chamber voltage drop (VR). (Radiation Protection) The ionisation chamber is the simplest gas-filled detector in which the electric current caused by the ion pairs created within the gas is measured. The X or gamma radia- tion passing within the gas interacts with its molecules through photoelectric, Compton or pair production process. The elec- Ionization S Vc trons accelerated by the electric field applied between electrodes chamber (Figure I.56) produce the ion pairs consisting of positive ions and C I free electrons. The intensity of the measured current depends on the exter- nal voltage value (Figure I.57). Its measurement is realised in – + the saturation region where all electric charge (ion pairs) is Vc ~ Exposure collected by the electrodes. The applied voltage is about 100 V and the ionisation current about 10−14 A. The guard rings are used to reduce the effects of the insulator leakage which would FIGURE I.59 Example of the ion current measurement with the use of a capacitor potential (VC). be added to the measured signal. Depending on the separation between the electrodes and the density of ion pairs in the gas some recombination of ion pairs in the gas may occur depending The ion current is measured indirectly by using either sensing on the voltage. of the voltage drop (Figure I.58) across the resistor (109–1012 Ω) or the voltage due to an electric charge gathered by the capacitor (Figure I.59). The integration method based on the measurement of current over a finite period of time can also be applied. The air-equivalent chamber is used for the exposure rate RE – E1 estimation as the ratio of the measured saturated current intensity X-ray + + + + – IS (A) to the mass m (kg) of the air in the active volume of the * – – – – + detector: E3 + E2 E3 RE = IS m (C/s/kg) Electrometer If the measurement is not performed at STP (T0 = 273.15 K, P0 = 1.013 × 105 Pa) the corrections for the air mass m must be applied. The ionisation chambers are used as radiation survey instru- FIGURE I.56 Scheme of an ionisation chamber: E1 – cathode, E ments. They can operate as sealed gas-filled detector or a gas flow 2 – anode, E3 – guard rings. counter to which a radioactive gas is provided to be measured. I Ionisation density 512 Ipsilateral Abbreviation: STP = Standard temperature and pressure. radiation must have sufficient energy to overcome the binding Related Articles: Absorbed dose, Exposure, Primary stan- energy of the electron that is emitted. Ionisation leaves the atom dard, Thimble chamber or molecule in a charged state. Further Readings: Dendy, P. P. and B. Heaton. 1999. Physics When the radiation does not have sufficient energy to cause for Diagnostic Radiology, Institute of Physics Publishing, ionisation then excitation will occur, this is when the radiation Philadelphia, PA, pp.137–139; Graham, D. T. and P. Cloke. excites the motion of an atom or molecule or induces an electron 2003. Principles of Radiological Physics, Elsevier Science to move from an occupied orbital state to a higher unoccupied Ltd., Edinburgh, UK, pp. 331–333; Knoll, G. F. 2000. Radiation state. Detection and Measurement, 3rd edn., John Wiley & Sons, New Particles (such as alpha particles and electrons) cause ioni- York, pp. 129–154. sation by electrostatic interactions. In tissue the average energy required to form an ion pair is about 34 eV. Ionisation density Electromagnetic radiation can cause ionisation by one of three (Radiotherapy) Ionisation density is defined as the number of processes: Compton scattering, the photoelectric effect or by pair ionisations per unit length along the track of a charged particle as production. The energy required depends on the atom or molecule it travels and deposits energy through a medium. This ionisation involved but generally electromagnetic radiation of more than a density increases along the charged particle’s path as the particle few electron volts is considered to be ionising radiation. This slows down until rapidly dropping off at the end of the range. means that only photons in the middle to far ultraviolet region, An increase in ionisation density is associated with an x-ray or gamma ray regions can cause ionisation. increased biological effect. This increased biological effect is Abbreviations: eV = Electron volt. thought to be caused by an increase in the number and complexity Related Articles: Auger particles, Beta particles, Compton of chromosome aberrations. scattering, Internal conversion electrons, Photo-electric effect, The macroscopic parameter linear energy transfer (LET) is Pair production often used to characterise ionisation density as it is relatively easy to calculate and achieves good correlation with biological effect. IORT However, it can be a poor parameter to describe ionisation density (Radiotherapy) See Intraoperative radiation therapy (IORT) on the micro- or nano- level. For instance, a proton of the same LET as a helium ion has a greater biological effect. This is due iPAT (integrated parallel acquisition technique) to the shorter range of the secondary electrons coming from the (Magnetic Resonance) Integrated parallel acquisition technique proton track than the helium ion track. This shorter range means (iPAT) is a vendor term (Siemens) for partial parallel imaging a greater ionisation density. (PPI) techniques, where determination of the coils’ sensitiv- Related Articles: Linear energy transfer (LET) ity profiles is integrated into the scan. The information for the so-called autocalibration is obtained from a specified number of Ionisation recombination loss central k-space lines representing a low-pass filtered signal. Since (Radiation Protection) The charged particles created by ionising these additional signals are eventually employed in the recon- radiation in gas-filled detectors may collide with neutral atoms structed image (e.g. in the GRAPPA algorithm), autocalibration I that are in thermal random motion. There are many types of col- implies a reduced loss of signal to noise ratio (SNR). lisions including charge transfer collisions. In the charge trans- Related Articles: Partial parallel imaging, GRAPPA fer collisions a positive ion can take an electron from the neutral molecule or a free electron can be attached to a neutral atom IPL classification (molecule). During the collision between the positive ion and the (Non-Ionising Radiation) See Intense pulsed light source free electron or a negative ion a neutral atom (molecule) may be formed. This process is called ion recombination. In such a situ- ation the electric charge is lost and it will not contribute in the Ipsilateral ionisation current measured by the detector. (Radiotherapy) Unilateral irradiation refers to restricting radia- There are two types of ionisation recombination loss: tion beams to the one laterality. This is distinctly different from a bilateral technique, in which both sides of the patient can be 1. Initial (columnar) recombination which occurs in a col- used for radiation beam positioning. If the unilateral irradia- umn of a track of an alpha particle or other strong ionis- tion is restricted to the same side as the treatment volume this is ing particles. then referred to as ipsilateral irradiation. Whereas the opposite 2. Volume recombination which occurs for ions and lateral position is referred to as the contralateral. The term ipsi- electrons moving towards the collecting electrodes. It lateral can be used to refer to the treatment technique or organ increases with increasing irradiation. To minimise this constraints. For example, ipsilateral radiotherapy can be benefi- effect it is necessary to apply a high electric field. cial in some head and neck cancers (e.g. carcinoma of the tonsil region). Ipsilateral can also be used to define an organ constraint; in lung radiotherapy, the dose to the ipsilateral lung will often be Related Article: Ion recombination reported. When delivering radiotherapy ipsilaterally a concern is Further Reading: Knoll, G. F. 2000. Radiation Detection and that there may be a contralateral failure in treatment. However, if Measurement, 3rd edn., John Wiley & Sons, Inc., New York, pp. this concern can be shown to be clinically insignificant, contralat- 131–132. eral sparing in some treatment sites can help preserve functional- ity. An indirect benefit of ipsilateral irradiation is that in the event Ionising radiation of requiring a treatment site to be re-irradiated, the non-irradiated (General) Ionising radiation is any form of radiation, particulate contralateral region can potentially be utilised to avoid some re- or electromagnetic, which can cause ionisation, which occurs irradiation of normal tissues. when an electron is emitted from an atom or molecule. The Related Articles: Contralateral Iridium-192 513 Iris diaphragm Iridium-192 Z (Radiotherapy, Brachytherapy) Iridium-192 is a reactor produced radionuclide (neutron activation of stable Iridium-191) with a half- life of 73.83 days. Iridium-192 decays via electron capture and beta emission to excited states of osmium-192 and platinum-192, Ø 0.9 which decay to stable states by gamma emission (x-rays are also produced). The photon spectrum of iridium-192 is complex, con- Center taining a large number of energy levels, with a maximum of more of the than 1 MeV. Iridium-192 is a photon source and the average photon sources energy is often given as 0.38 MeV for an encapsulated high dose rate source (the exposure weighted effective energy is 397 keV). The decay

beta radiation is mostly absorbed in the encapsulation. Y It is to be noted, that secondary photons and electrons are produced in photon interactions both in the source material and the encapsu- Iridium-192 lation; care must be taken in source strength measurements. core Ø 6 Ø 0.7 Types of Iridium-192 Sources: Iridium-192 is widely used for temporary implants, and there are sources available for most types of brachytherapy; intracavitary and interstitial, high dose Laser rate, low dose rate and pulsed dose rate techniques. Iridium is welded not used for permanent implants due to the rather high averages photon energy causing radiation protection problems (compared to the low photon energy of iodine-125.) Steel Low dose rate sources: Iridium sources are available as plati- cable num coated wires, and as small stranded seeds. These sources are generally handled with afterloading techniques, both manu- ally and remotely controlled. (The Paris system for interstitial implants was designed for low dose rate iridium wires.) High dose rate and pulsed dose rate sources: Iridium-192 is well suited for high dose rate sources due to its high specific FIGURE I.60 High dose rate source for the GammaMed Plus afterload- source strength. Small high intensity sources make it possible to ing unit (Varian). use small applicators together with short treatment times. Typical source strength for a high dose rate source is 40.70 mGy/h at 1 m (370 GBq = 10 Ci apparent activity), and sources with sizes in the order of Φ 0.7 mm and length 3.5 mm, encapsulation Φ 0.9 mm Innenlange ausfahrschlauch + applikator 1300 –0.5 and length 4.5 mm, are available, see the drawing in Figure I.60. –0.1 Inside length source guide tube + applicator 0.5 3.5 +1 High dose rate and pulsed dose rate sources can only be used –1 with remote afterloading units. The high dose rate source in Figure Variabel je 2.5 I Gerundet/rounded I.60 (GammaMed Plus, Varian) is welded to a drive cable, and nach applikator depends on type source stop positions and dwell times are controlled by the precision of applicator drive motor in the afterloading unit and the accompanying treat- Strahlerkabel/source cable ment control software. Figure I.61 shows the position of the source at its first, most distal, stop position in an applicator, together with a marker wire mimicking stop positions with a step size of 1 cm. Due to the relatively short half-life of iridium-192, the source is usually replaced four times per year, to keep treatment times short. Each source exchange requires extra quality control tests besides the ordinary daily tests, and determination of source strength and determination of source stop positions are the two most important parts of the specific source exchange QA programme. Marker Related Articles: Brachytherapy sources, Remote afterload- ing units 2.75 1.5 3.5 Iris diaphragm (Diagnostic Radiology) The iris diaphragm is normally placed in FIGURE I.61 The first, most distal, stop position of this source in an front of the TV camera of a digital fluoroscopic system (such as applicator. Also shown is a marker wire inserted in the applicator, indi- DSA). The device stays between the image intensifier output and cating source stop positions with a step length of 1 cm. the TV camera (it is similar to the aperture of a normal photo- graphic camera). The iris diaphragm provides exact light beam size, hence overall light intensity, to the TV camera input/target. The Iris diaphragm influences also the signal to noise ratio The adjustment of the iris diaphragm is related to the proper (SNR) of the final image. If the aperture of the Iris diaphragm functioning of the TV camera tube (such as Vidicon type), avoid- is smaller less light from the Image Intensifier will reach the TV ing overexposure (saturation) or underexposure of its target. This camera. This could drive the automatic feedback system (auto- way the TV camera will always function within the linear part of matic brightness control) to increase the radiation dose, hence its characteristic. the overall light intensity. This will increase the patient dose but 0.3 0.62 3.5 0.1 4.52 Iron-based contrast agents 514 Irradiance at the same time will decrease the quantum noise and produce image with better SNR. Related Articles: Automatic brightness control, Digital sub- traction angiography (DSA), x-ray television, Vidicon TV camera Iron-based contrast agents (Magnetic Resonance) MRI contrast agents are contrast agents used to improve the visibility of internal body structures in mag- netic resonance imaging. Gross tumour volume Two types of iron oxide contrast agents exist: Clinical target volume Planning target volume • Superparamagnetic iron oxide (SPIO) • Ultrasmall superparamagnetic iron oxide (USPIO) Treated volume Irradiated volume These contrast agents consist of suspended colloids of iron oxide nanoparticles and when injected during imaging reduce the T2 FIGURE I.62 Definition of target volumes as in ICRU 50. signals of absorbing tissues. SPIO and USPIO contrast agents have been used successfully for liver tumour enhancement or MR angiography. and inter-site comparability. It details the minimum set of data After injection, they accumulate in the reticuloendothelial sys- required to be able to adequately assess treatments without having tem (RES) of the liver (Kupffer cells) and the spleen. to return to the original centre for extra information (Figure I.62). At low doses, circulating iron decreases the T1 time of blood, Related Articles: Planning target volume (PTV), Clinical tar- at higher doses predominates the T2* effect. get volume (CTV), Treated volume, Gross tumour volume (GTV), Iron-based agents are much more effective in MR relaxation Parallel organ, Serial organ, Tolerance than paramagnetic agents. The contrast between liver and lesion Further Readings: ICRU. 1993. Prescribing, reporting and improves because hepatic tumours either do not contain RES cells recording photon beam therapy, ICRU Report 50, International or their activity is reduced. With all standard pulse sequences, super- Commission on Radiation Units and Measurements, Washington, paramagnetic iron oxides cause noticeable shorter T2 relaxation DC; ICRU. 1999. Prescribing, recording and reporting photon times with signal loss in the targeted tissue (e.g. liver and spleen). beam therapy (Supplement to ICRU Report 50), ICRU Report 62, Use of these colloids as tissue-specific contrast agents is now a International Commission on Radiation Units and Measurements, well-established area of pharmaceutical development. Washington, DC. Ferid ex®, Endor em™, Gastr oMARK ®, Lu mirem ®, Si nerem ® and Resovist® are some of the most commonly used makes. Irradiance Some remarkable points about these agents: (Non-Ionising Radiation) It is the most common term used in light therapy or safety to indicate the light received by the stricken I • A minimum delay of about 10 min between injection subject/object. This is the light flux through a unit area, and it is (or infusion) and MR imaging extends the examination measured in Wm–2. time Skin exposure to solar or artificial light is usually expressed • Cross-section flow void in narrow blood vessels may as an irradiance. impede the differentiation from small liver lesions • Aortic pulsation artefacts become more pronounced Related Articles: Super param agnet ism, Super param agnet ic Co ntras t Age nts Further Readings: Shen, Z., A. Wu and X. Chen. Iron oxide nanoparticle based contrast agents for magnetic resonance imaging; Xiao, Y.-D., R. Paudel, J. Liu, C. Ma, Z.-S. Zhang and S.-K. Zhou. MRI contrast agents: Classification and application (Review). IRPA (General) See International Radiation Protection Association Irradiated volume (Radiotherapy) The irradiated volume in radiotherapy is the tis- sue volume that receives a dose that is considered significant in relation to normal tissue tolerance. The significance level will depend on the normal tissue type. For more information about significant dose please see the articles on Parallel organ, Serial organ, and Tolerance. The use of ‘irradiated volume’ as a planning volume was proposed by the ICRU in Report 50 (with addendum 62). This report provided a common framework on prescribing, record- ing and reporting therapies, with the aim to improve consistency Irregular field 515 Isodose curve Related Articles: AORD, Light source, Photodiode, y Phototherapy, Radiance, UV dosimetry Further Readings: Czapla-Myers, J. S., K. J. Thome and S. F. Biggar. 2008. Design, calibration, and characterization of a field x radiometer using light-emitting diodes as detectors. Appl. Opt. 47(36):6753–6762; Ihrke, I., J. Restrepo and L. Mignard-Debise. 2016. Principles of light field imaging: Briefly revisiting 25 years of research. IEEE Signal Process. Mag. 33(5):59–69; Kitsinelis, S. and S. Kitsinelis. 2015. Light Sources: Basics of Lighting z Technologies and Applications. CRC Press, Boca Raton, FL. Irregular field (Radiotherapy) This is a term generally given to an open field that has been shielded using low melting point alloy, lead or multileaf collimators (MLC) (Figure I.63). Related Articles: Multileaf collimator, Equivalent square FIGURE I.64 The coordinate system of a cylindrical magnet. ISIS Isocentre (Magnetic Resonance) See Image selected in vivo spectroscopy (Radiotherapy) This is the ‘point’ about which all the various (ISIS) axes of rotation of the treatment machine meet. The gantry of the accelerator, the collimators, and the couch all rotate about this Isobars point. The room lasers are also aligned to meet at this position. (General) Isobars are nuclides with the same mass number; that is There are also two types of isocentre – mechanical and radiation. the same number of nucleons (protons and neutrons). The nuclides The mechanical isocentre relates to the point about which all the will therefore be of different elements. equipment rotates about, while the radiation isocentre is the point At lower mass numbers normally only one of the isobars is sta- about which beams delivered from different gantry angles meet. ble – an example, N-14 is stable but all the other isobars of N-14, Ideally these points should be the same. e.g. C-14, are radioactive. However, with increasing mass there It is becoming increasingly common for treatments now to be are more examples of several isobars being stable – an example: delivered isocentrically – that is with the centre of the tumour Ar-40 and Ca-40 are all stable. K-40 is not stable but has a very positioned at the isocentre, rather than with a fixed SSD. long half-life. Related Articles: Mechanical isocentre, Radiation isocentre, Related Articles: Isotope, Isotone, Nucleons Isocentric technique Isocentre Isocentric technique (Magnetic Resonance) The point at the centre of the magnet of (Radiotherapy) This technique typically places the centre of an MR scanner is called the isocentre. It defines the origin (x, the target at the isocentre position for all treatment fields, and I y, z = 0,0,0) of a right handed coordinate system (Figure I.64). so the source to skin distance (SSD) will change for different This coordinate system can be used to describe the directions beams depending on the depth of tissue the beam must penetrate. of the gradient fields (see related article) as well as many other Another name for this technique is the fixed source axis distance parameters related to pulse sequences, the position of the (SAD) technique. patient and geometrical properties of the MR images. Also, the The alternative technique to this is to maintain a constant SSD isocentre is located at the centre of a usually spherical or ellipti- to the skin surface and so the target will be at varying distances cal volume with a very homogeneous B0 field. This volume is for each field. where the imaged part of the patient is located during an MR Abbreviations: SAD = Source axis distance and SSD = Source examination. B0 as well as the resonance frequency of the MR to skin distance. scanner are defined only at the isocentre. Thus, the isocentre is Related Articles: Isocentre, Multiple isocentre treatment, not a point in space but represents the maximum useable field- Source axis distance (SAD), Source-to-skin distance (SSD) of-view (FOV). Related Articles: Gradient coils, Gradient fields, Field-of-view Isochromat (Magnetic Resonance) An isochromat is the line that connects all the point in the object plane in which a group of spins have equal phase evolution during a segment TR of a Fast Field Echo Regular field Irregular field sequences. The shape of an isochromat depends on the gradients used in the imaging sequence and also on inhomogeneity of the Irradiated area main magnetic field and the chemical shift effect. In practice a Irradiated area group of protons may have a range of resonance frequencies and we call isochromat each group of nuclear spins that share the same resonance frequency. Isochromats have a smaller dimension than

magnetic resonance voxels. Shielded areas Isodose curve (Radiotherapy) An isodose curve is a line of constant absorbed FIGURE I.63 Regular and irregular fields. dose. The isodose curves are generally drawn at regular intervals Isodose rate surface 516 Isoeffect Distance from central axis (cm) Further Readings: ICRU. 1985. Dose and volume specifica- –10 –8 –6 –4 –2 0 2 4 6 8 10 tion for reporting intracavitary therapy in gynaecology, ICRU 0 Report 38, International Commission on Radiation Units and Measurements, Washington, DC; ICRU. 1997. Dose and volume specification for reporting interstitial therapy, ICRU Report 58, International Commission on Radiation Units and Measurements, 90 Washington, DC. –5 80 Isodose shift method (Radiotherapy) The isodose shift method is a simple method to correct for oblique beam incidence or non-uniform patient con- 70 tours across the beam. In the oblique entry situation, the dose dis- tribution will be different to that as measured for a flat phantom –10 surface with a beam incident at the normal angle to the surface. 60 In this method the value of the dose at a point P is shifted on a vertical line by a quantity (k × h), where k is a factor to take account of the energy of the beam used and h is a measure of the thickness of tissue either missing or in excess at that point relative 50 to the central axis (Figure I.66). –15 In situations where there is missing tissue, the quantity h will be a positive value and so the isodose lines are shifted deeper into the patient or phantom, while the opposite is true for excess tissue 40 where h has a negative value. 30 The factor k is always less than 1 and is 0.7 for cobalt beams 20 10 and energies less than 5 MV, 0.6 for energies between 5 and 15 –20 MV, 0.5 for energies between 15 and 30 MV, and 0.4 for energies greater than 30 MV. Related Articles: Oblique incidence, Obliquity, Obliquity FIGURE I.65 Isodose curves. effect Further Readings: Podgorsak, E. B. 2003. Review of Radiation Oncology Physics: A Handbook for Teachers and of absorbed dose and are expressed as percentage of the dose at a Students, International Atomic Energy Agency, Vienna, Austria; normalisation point (i.e. 80%, 70%, etc.). A set of isodose curves Williams, J. R. and D. I. Thwaites. 2000. Radiotherapy Physics in is called isodose chart. These refer commonly to principal planes Practice, 2nd edn., Oxford University Press, Oxford, UK. that are planes which contain the beam axis. Measurements of the I isodose curves should be made in a water tank large enough to Isodose surface permit a full scatter condition to the point at which measurements (Radiotherapy) Isodose surfaces are a family of surfaces in 3D are being made or in a tissue equivalent phantom. In Figure I.65 which connect all points of equal dose in an irradiated volume isodose curves are shown for a normal incidence of the photon as shown in Figure I.67. A 2D plane taken through these surfaces beam on a homogeneous phantom. The isodose curves could be produces a family of isodose curves or isodose lines as shown for measured by an ionisation chamber, a solid-state dosimeter, ther- a lung patient in Figure I.68. moluminescence dosimeters and films. A precision carriage could be mounted within the water tank allowing the 3D placement of Isoeffect the detector at any position in the tank under the remote control (Radiation Protection) The isoeffect approach can be used to of a computer. study combined effects of mixed radiations. In principle it is the Isodose rate surface (Radiotherapy, Brachytherapy) Volumes in Brachytherapy: An isodose surface is a surface that defines a volume enclosed by that surface. Similarly an isodose rate surface is a surface that defines a h2 volume enclosed by that surface. Isodose and isodose rate surfaces are used in both permanent h1 and temporal implants at low dose rate. For high dose rate and pulsed dose rate implants, only isodose surfaces as used. Examples: the reference volume of ICRU Report 38 is the vol- ume enclosed by the reference isodose surface. ICRU Report 58 Point P1 Beam Point P defines the reference volume as the volume encompassed by an 2 central isodose surface in relation to the mean central dose. axis Abbreviation: ICRU = International Commission on Radiation Units and Measurements. FIGURE I.66 A simple illustration showing an oblique surface and how Related Articles: Reference isodose, Reference volume, Paris the quantity h can be measured for missing tissue (point P1, left, a positive system, Manchester system h value) and excess tissue (point P2, right, a negative h value). Depth (cm) Isomeric transition (IT) 517 Isomers nuclide is in an excited nuclear state. In most cases the nucleus then de-excites virtually instantaneously, in less than a picosec- ond, with the emission of gamma radiation and the end result is the daughter in its ground state. However, in a small number of radionuclides the excited daugh- ter has a relatively long lifetime – measured in seconds, minutes or hours. These long-lived excited states are called ‘metastable levels’ and are designated by the use of the suffix or superscript m (for metastable). An example is the long-lived excited state of technetium-99, which is designated technetium-99m (99Tcm). Particulate radiation is only emitted in the form of electrons as a result of internal conversion of the gamma radiation. Technetium-99m is widely used in nuclear medicine and is the excited daughter of the radionuclide of molybdenum – see Figures I.69 and I.70. Related Articles: Internal conversion, Gamma radiation, Radioactive decay, Radionuclide FIGURE I.67 Isodose surface through a prostate patient case. Isomers (General) In physics, an isomer is each of two or more atomic nuclei with the same atomic number and mass number but dif- ferent energy states. Derived from the Greek isomerês – ‘sharing equally’. The term nuclear isomer is often used to distinguish it from other types of isomers. A particular type of nuclear isomer is when a nucleus is in an excited metastable state which has a relatively long lifetime – measured in seconds, minutes or hours. An isomeric transition 99 42Mo 66.7 h β– 1 β– 2 921 keV Molybdenum-99 I β– 3 γ8 FIGURE I.68 Isodose lines in one plane through the case in Figure I.67. 509 γ7 γ6 same method used to study the effect of drugs and antibiotics. The 181 γ4 method consists of in vitro and in vivo measurements. For beams γ1 γ3 γ 142.6 5 140.5 containing radiation components belonging to a range of linear 0 γ2 energy transfer (LET) it is important to study the interaction of 99 43Tc Radioactive 2 × 105 y these radiation components of different qualities (crucial in the planning of the radiotherapy treatment with neutron and charged particles). Studies have demonstrated that, in certain cases, the FIGURE I.69 Decay of Mo-99 to Tc-99m and Tc-99. combined effects of high and low LET are greater than what could be calculated considering the independent effects of each compo- nent. There are models to evaluate the synergism of radiations of 99 different LET. There are also showing that the simple isoeffect 43 Tcm 6 h γ1 relation is applicable to a certain class of radiobiological data. A 142.6 keV detailed investigation is required for each particular combination 140.5 of radiations. Technetium-99 m Hyperlink: https//www .IAEA .org γ3 γ2 Isomeric transition (IT) (General) An isomeric transition is the transition of a nucleus in an excited metastable state to the ground state with the emission 0 99Tc Radioactive 2 × 105 of a gamma ray. 43 y A nucleus in an excited metastable state is called a nuclear iso- mer. After radioactive decay of many radionuclides, the daughter FIGURE I.70 Decay of Tc-99m to Tc-99. Isotones 518 Iterative image reconstruction is the transition of a nucleus in an excited metastable state to the reconstruction methods (such as FBP) is that it is relatively easy ground state with the emission of a gamma ray. to incorporate prior knowledge into the reconstruction process in Related Articles: Isomeric transition, Radioactive decay order to better handle noise or compensate artefacts. Rather than assuming a continuous formulation of the prob- Isotones lem (as done for the FBP), these methods consider the discrete (General) Nuclides are isotones if they have the same number of (or ‘digital’) nature of the acquired and reconstructed images. neutrons in the nucleus. The nuclides may be stable or radioactive. Consider the acquired projection data (sinogram) arranged as a Helium ( 4 2 He) has two protons and two neutrons but deuterium singular column vector: ( 2 1 D) also has two neutrons and so these two nuclides are isotones. The term isotone is derived from the term isotope (nuclides p = (p1,…, p T M ) having same number of protons) by replacing the p with an n. Related Articles: Isotope, Nuclide and, similarly, the image to reconstruct as a singular vector Isotope T (General) An isotope is one of a group of nuclides that have f = ( f1,…, fN ) the same proton (atomic) number. This group is composed of different nuclides of the same element that differ in neutron According to the number and size of each detector column, the number and, in some cases, nuclear state. Because they differ number of acquired projections and the desired pixel size of the only in their complement of neutrons they have similar physi- reconstructed image, an M×N weight matrix (also called system cal properties and the same chemical properties. Isotopes are matrix) can be defined as either stable or unstable (radioactive) in which case they may be referred to as radioisotopes. Radioisotopes undergo radioac- æ a11 a12 … a1N ö tive decay with a specific half-life. In general, the greater the ç ÷ difference in mass between an unstable and a stable isotope, A = ç a21 . a2N ÷ the greater is the instability of the radioactive isotope and the ç . ÷ ç shorter the half-life. ç ÷ a ÷ è M1 aMN ø Related Articles: Radioisotope, Radionuclide, Half-life, Radioactive decay and its values result from geometrical considerations represented as Isotropic (General) For isotropic radiation the intensity is uniform in every direction, i.e. not dependent on the direction. For example, the particles and photons emitted during decay are emitted isotropi- cally. The antonym for isotropy is anisotropy, i.e. the radiation intensity is direction dependent. I Isotropic emission (Nuclear Medicine) The intensity from a radioactive source with isotropic emission is independent of radiation direction, i.e. the source has no preferred direction of radiation. Isotropic emission is one of the fundamental physical rules and applies to all kinds of radioactive decay. Even though the emission from a single radio- active nucleus or small sample is isotropic the emission from a radioactive sample seldom is because of sample spread and self- absorption within the sample. The algebraic formulation can be finally modelled as the Isowatt circuit matrix product: (Diagnostic Radiology) See Automatic brightness control (ABC) p = Af Iterative algorithm (Nuclear Medicine) See Iterative reconstruction methods An exact solution of the resulting system of linear equations is not possible due to the unavoidable noise in the acquired projec- Iterative image reconstruction tion images. Iterative methods basically differ for their approach (Diagnostic Radiology) In projection reconstruction techniques to solve the above system of equation and/or for the relaxation like x-ray computed tomography (CT), iterative reconstruction parameters used as a trade-off between the convergence of the methods are an alternative to the most widely adopted recon- system (which affects computational time) and the signal to noise struction technique named filtered back-projection (FBP). These ratio of the output reconstructed image. Examples of iterative methods have been discarded up until now as being too compu- methods are the algebraic reconstruction technique (ART) and tationally intensive, but with increased computer speeds (espe- the simultaneous algebraic reconstruction technique (SART). cially with GPU devices), these methods are becoming a more Related Articles: Filtered back-projection (FBP), Signal-to- common alternative. An advantage of these methods over other noise ratio, Algebraic reconstruction technique (ART) Iterative reconstruction methods 519 IUPESM Iterative reconstruction methods MLEM (Maximum-Likelihood Expectation-Maximisation): (Nuclear Medicine) The iterative reconstruction method is an This algorithm will produce the ‘most likely’ (maximum likeli- alternative to filtered backprojection (FB), but has until recently

hood) source distribution derived from the observed projections. been discarded as too computationally intensive. But as computer In this reconstruction method all projection angles are used to speed has increased the method is becoming a more common update the estimated image. This procedure slows down the pro- alternative. cess and it requires several iterations before a reasonable image is The reconstruction starts with an estimation of the true acquired. The time consumed in the backprojection during each image. This estimation is often a simple image, such as a blank iteration is proportional to the number of projections used. OSEM or uniform image. The algorithm will eventually approach the is a commonly-used algorithm used that has found a way around truth by performing gradual approximations, or iterations. The this problem. next step is to acquire the projection that would have been mea- OSEM (Ordered Subsets Expectation-Maximisation): sured to attain the estimated image. This process is an inverse Instead of using all the projections to estimate an image, the backprojection, or forward projection and it is basically a sum- OSEM algorithm only uses a subset of projections, meaning four mation along the potential ray paths for each projection. The orthogonal projections. The projections are used to calculate a projections from the estimated image are then compared to the new estimated image. The procedure is repeated with four new measured projections. The two sets of projections are not likely projections to make a new estimated image. This algorithm is to agree at an early stage since the true image differs from the faster and produces images with high spatial resolution after one initial estimation image which is a blank or uniform image. But iteration, unlike MLEM where the image resolution is enhanced the difference between the two projection sets can be used to for every iteration using all projections. update the estimation image. This update process is repeated To get a more elaborate explanation of iterative image recon- until the similarity between the two images reaches a certain struction methods the reader is referred to respective chapters in threshold. the following two references. As described earlier, the concept of iterative reconstruction Further Readings: Cherry, S. R., J. A. Sorenson and M. E. algorithms consists of two basic components: (1) the method Phelps. 2012. Physics in Nuclear Medicine, 4th edn., Elsevier for comparing the true projections and the estimated and (2) the Saunders, Philadelphia, PA, pp. 253–277; Wernick, M. N. and J. method where the estimation is updated according to the compar- N. Aarsvold. 2004. Emission Tomography. The Fundamentals of ison. Using iterative reconstruction methods it is possible to give PET and SPECT, Elsevier, London, UK, pp. 443–448. different weights to different data in the projection. Specifically, Related Articles: Filtered backprojection, Signal-to-noise give higher weight to high count elements and less weight to low ratio count elements to increase the signal-to-noise ratio. Other advan- tages when using iterative reconstruction methods is that it is pos- IUPESM sible to model physical properties, like scatter events and septal (General) See International Union for Physical and Engineering penetration. Sciences in Medicine I J J coupling It is the minimum amount by which stimulus intensity must (Magnetic Resonance) An alternative term for scalar magnetic be changed in order to produce a noticeable variation in sensory coupling, in which J is the coupling constant characterising the experience (University of South Dakota, n.d.). interaction between nuclei. In other words, JND is defined as the difference in luminance Related Article: Magnetic coupling required to be visibly noticed by an observer under specific view- ing conditions (AAPM, 2019). JND (Just noticeable difference) JND is useful when measuring the luminance change in pri- (General) See Just noticeable difference (JND) mary reporting monitors where a change in luminance must be calibrated to the DICOM (Digital imaging and communications Joule in medicine) greyscale standard. When assessing the ability of a (General) The joule (symbol J) is the unit of energy (work, heat) monitor, the change in luminance is often measured against this in the International System of Units (SI), equal to the product of standard. It is these results that indicate whether the monitor is the force in newtons (SI unit – symbol N) and the distance in satisfactory for clinical use. metres (SI unit – symbol m). JND index: The input value to the greyscale standard display The Joule is therefore a derived SI unit that can more properly function (GDSF), such that one step in the JND index results in a be defined in terms of base SI units, expressed as luminance difference that is just noticeable. 1J = 1kg ´ (m2 /s2 For a certain change in JND, the perceived change luminance ) is greater at low luminance level than at high luminance lev- where els. This means that at high luminance levels, a greater change kg is kilogram (mass) in luminance is required in order to recognise a perceived m is metre (distance) change for the average human observer. This is illustrated in s is second (time) Figure J.1. Related Articles: DICOM (Digital imaging and communica- The Joule is named after the English physicist James Prescott tions in medicine), Image display, Human visual response func- Joule (1818–1889). tion, Greyscale standard display function, Medical image display, Related Article: SI (Système International) Barten model Further Reading: Benedek, G. B. and F. M. H. Villars. Further Readings: AAPM, 2019. Display Quality Assurance 2000. Physics with Illustrative Examples from Medicine and Report 270, AAPM; Fetterly, K. A., Blume, H. R., Flynn, M. J. Biology: Mechanics, 2nd edn., Springer-Verlag Inc., New York, and Samei, E., 2008. Introduction to grayscale calibration and pp. 354–355. related aspects of medical imaging grade liquid crystal displays. J. Digit. Imaging 21(2): 193–207; University of South Dakota, n.d. Justification Weber’s Law of Just Noticeable Differences. [Online] Available (Radiation Protection) The first principle of protection against at: http: / /app s .usd .edu/ cogla b /Web ersL a w .htm l (accessed 16 July J ionising radiation for workers, patients and members of the pub- 2019). lic, specified by the International Commission on Radiological Hyperlink: Weber’s Law – http: / /app s .usd .edu/ cogla b /Web Protection (ICRP), is justification. ersL a w .htm l Any practice involving the exposure of staff, patients or the public to ionising radiation must be justified – a process that evalu- ates the benefits to the exposed individuals or to society that result from such a practice against the risks associated with the potential adverse radiation effects of the exposure. The most recent ICRP Recommendations (ICRP 103, 2007) has the following definition: ‘The Principle of Justification: Any decision that alters the radia- tion exposure situation should do more good than harm’. Once a practice has been justified, then the second and third principles must be applied – i.e. Optimisation and Dose limitation. Related Articles: Adverse radiation effects, Dose limitation, International Commission for Radiological Protection (ICRP), Optimisation Further Reading: The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann. ICRP 37(2–4). Just noticeable difference (JND) FIGURE J.1 DICOM greyscale standard display function plotted as the (General) Just noticeable difference (JND) is a metric used to logarithm of the luminance as a function of JND. (Courtesy of East and measure the luminance change. North Hertfordshire Trust, Radiation Protection Department, 2019.) 521 K K absorption edges represents the kinetic energy transferred to charged particles per (General) When the energy of an incoming photon exactly unit mass of irradiated air when indirectly ionising (uncharged) matches that required to eject a K-shell electron, strong resonant radiations such as photons traverse the volume of air. photoelectric absorption occurs, producing what is known as an Kerma is the starting point for determining the energy depo- absorption edge. sition by a given type of radiation in an absorbing medium and Wavelengths shorter than the K absorption edge have suf- varies according to radiation type and physical characteristics of ficient energy to eject K orbital electrons; wavelengths slightly the medium itself. longer than the absorption edge do not. As photon energy is Kerma is formally defined as the kinetic energy released per increased beyond the absorption edge, the interaction cross sec- unit mass: tion decreases. K absorption edges have a number of important applications dE in diagnostic radiology. K = dm The energy at which the K absorption occurs is determined solely by the atomic number of the absorber. This means that where dE is the sum of the kinetic energies of all the charged sequences of materials with adjacent atomic numbers can be used particles liberated by uncharged ionising radiation in a mass dm. to filter photon beams, permitting transmission of only photons This includes the energy of any Auger electrons. within a narrow band of energies. Following the absorption of The units of kerma or air kerma are J/kg or Gray (Gy). photons at the K-edge, characteristic x-rays are emitted. These As an approximation, the absorbed dose can be related to the will not be absorbed by the emitting material (their energy is kerma taking into account the fraction of the energy released in slightly below the K-edge), but can be absorbed by materials with the mass, which is not absorbed within that mass: slightly lower atomic numbers. This principle is used to ‘harden’ beams generated by x-ray tubes, using combinations of copper, Dair » K (1- g) aluminium and bakelite filters. Absorption edges also provide the fundamental mechanism where g is the fraction of energy of the charged particles liberated behind x-ray contrast media. Radio-opaque contrast media such by the photons, which is lost in radiative processes – i.e. where as iodine and barium have K absorption edges matched to the particles are not fully stopped within the mass. At lower energies, spectrum of x-rays leaving the patient. g is relatively small, implying that transient charged-particle equi- Related Articles: Absorption cross section, Absorption by librium has been quickly reached, whereby the energy of charged contrast media, Filter, Photoelectric absorption particles exiting the volume is equal to the energy of charged particles entering the volume from forward and backscatter. As K-edge the energy of the incident radiation increases, the distance into (General) See K absorption edges a medium that this equilibrium is reached becomes deeper. This depth is of significance in radiotherapy treatment planning, and is K-edge metal filter called the build-up region. (Diagnostic Radiology) The K-edge filter is used to attenuate the The relationship between absorbed dose and kerma at depth high photon energy range of an x-ray spectrum. It is based on the within the medium may be described by the graph in Figure K.2. principle that a material produces maximum x-ray attenuation by Related Articles: Absorbed dose, Transient charged-particle the photoelectric process for photon energies just above the bind- equilibrium, Energy deposition K ing energy of the K-shell electrons. The energy of the K-edge (K-shell electron binding energy) is Kernel determined by the atomic number (Z) of the material. Therefore, (General) Kernel is an image processing term associated with a filter material can be selected to adjust the K-edge attenuation to filtering. In order to apply a filter to an image, a kernel matrix the desired photon energy as illustrated in Figure K.1. (containing multiplication factors) is applied to every pixel in the Molybdenum (Z = 42) is the most common K-edge filter mate- image. Each pixel and its neighbours are multiplied by these fac- rial and is used in mammography to attenuate the spectrum above tors and then the pixel is replaced by the sum of the products. its K-edge energy of 20 keV. Rhodium (Z = 45) is an alternative This operation is mathematically known as convolution. It is filter material with a K-edge energy of 23.22 keV. Switching from defined as follows: the molybdenum to the rhodium filter extends the x-ray spectrum, making it more penetrating and more useful in the imaging of ¥ dense breast. g ( x) = òI ( x¢)h(x - x¢)dx¢ Related Articles: K-edge, Filtration total, X-ray beam filtration -¥ Hyperlink: http://www .sprawls .org /resources /MAMMO where Kerma I(x) is the input image (Radiation Protection) Kerma (kinetic energy released per unit g(x) is the result mass) is used to describe energy loss in a medium. Thus, air kerma h(x) is the convolution kernel 523 Kernel-based treatment planning

524 Keyhole imaging e K-edge Equation K.1 shows the superposition equation for general calcu- lation of dose at a point r. In Equation K.1, the product of the mass attenuation coefficient Filter and the primary energy fluence, denoted by A, is the TERMA (total energy released per unit mass), which represents the dose to the irradiated medium by the interaction of primary photons, and which can be readily calculated by ray tracing through the medium in a similar way to determining the radiological depth. The term B is referred to as the kernel and describes the pat- tern of spread of energy away from the interaction point. Each point in the kernel describes the dose deposited, by scattered pho- tons and electrons set in motion, as a fraction of the TERMA at the interaction point. These kernels can be obtained from decon- 0 20 40 60 80 120 volution of dose distributions or by direct measurement, but the Photon energy (keV) most common approach is to generate the kernels using Monte Carlo techniques (Mackie et al., 1988). The kernels can take the FIGURE K.1 A K-edge filter that attenuates the x-ray photon energies form of point, pencil or planar kernels and can be further decom- above the K-edge energy. (Courtesy of Sprawls Foundation, www .sprawls posed into contributions from primary dose, single scatter dose .org) and multiple scatter dose. In the special case of spatially invariant kernels, Equation K.1 simplifies to a convolution integral where K(r − r′) replaces K(r, r′). The integral can then be efficiently evaluated using Fast Fourier Transforms, at a cost of reduced accuracy compared to Gy the superposition integral. Kerma The convolution–superposition class of dose calculations have been widely implemented into commercial treatment planning systems using point kernels, pencil beam kernels, dose deposi- Absorbed dose tion kernels and collapsed cone approximations (Figure K.3). Heterogeneities are accounted for by scaling the kernels for dif- ferent tissues. Polyenergetic beams are accounted for by using a weighted sum of monoenergetic kernels. Further Readings: Ahnesjö, A. and M. Aspradakis. 1999. Dose calculation for external photon beams in radiotherapy. Phys. Build-up Depth in tissue Med. Biol. 44:R99–R155; Mackie, T. R., A. F. Bielajew, D. W. O. Rogers and J. J. Battista. 1988. Generation of photon energy deposition kernels using the EGS4 Monte Carlo code. Phys. Med. FIGURE K.2 Attenuation of photons in tissue showing build-up region. Biol. 33:1–20; Ostapiak, O. Z., Y. Zhu and J. Van Dyk. 1997. Refinements of the finite-size pencil beam model of three-dimen- sional photon dose calculation. Med. Phys. 24:743–750. The convolution operation is usually denoted by *: g ( x) = I ( x)* h( x) = h( x)* I ( x) Keyhole imaging (Magnetic Resonance) Keyhole imaging is a technique to speed Instead of using convolution to apply a filter to an image, it is up the acquisition of dynamic sequences in MRI. Normally, the entire k-space for each image is collected in a K much easier to use Fourier transforms. This is because the Fourier transform of the convolution of two functions is equal to the prod- dynamic sequence. However, much k-space information in each uct of their individual Fourier transforms: subsequent image is identical in dynamic sequences; for example the information about the vessel edges in an angiography imag- FT {I ( x)* h( x)} = FT {I ( x)}´ FT {h( x)} ing sequence is likely to remain the same throughout the imaging time, if the vessel is not moving. Related Articles: Filtering, Spatial filtering This is utilised in the keyhole technique. Instead of collecting all k-space phase encoding lines in each image, (morphological) Kernel-based treatment planning information from a reference image is used instead. (Radiotherapy) In model-based approaches to dose calculation, The central part of k-space holds most of the contrast informa- the dose distribution is modelled from first principles. One such tion in an image. In contrast, the outer parts of k-space depict the approach involves the convolution–superposition class of meth- distinct borders in the image, such as vessel walls. When imaging, ods, in which the primary photon interactions are dealt with sepa- for example contrast agent going through vessels, a complete image rately from the transport of scattered photons and electrons set in is acquired in the beginning before the contrast injection. During motion. In this approach, the dose at any point in an irradiated the contrast injection, only the central lines of k-space in the volume, D, is expressed by Equation K.1: phase-encoding direction are collected. The missing outer parts of k-space are filled with data from the reference image (Figure K.4). A  In this way, information about the vessel signal (the contrast) is B  ( ) ò m collected and updated during the examination, while information D r = Y(r¢)K (r ,r¢)d3r¢ (K.1) r about vessel borders is taken from the reference image. Attenuation coefficient (per mm) Kit 525 Kit –30 –20 –10 0 10 20 30 1.2 0 1.0 5 0.8 10 0.6 15 0.4 = 20 0.2 25 30 0.0 200 150 35 100 200 40 50 150 100 0 50 45 0 Energy fluence Dose deposition Absorbed dose kernel FIGURE K.3 Example of determining a dose distribution from convolution of a dose deposition kernel with energy fluence. (Courtesy of Tommy Knöös, Lund University Hospital, Lund, Sweden.) Reference image . Dynamic sequence FIGURE K.4 Keyhole illustration. Prior (or after) the dynamic sequence, a reference image is acquired of outer k-space lines. In the dynamic sequence, the central parts of k-space in each image are acquired and missing data is taken from the reference image. The keyhole technique is often used to increase the tempo- The preparation of the radiopharmaceutical is performed by ral resolution when imaging dynamic processes. However, organ the injection of a sterile solution of 99Tcm-pertechnetate to the motion should be avoided in order not to compromise spatial kit vial by using aseptic techniques. A breather needle should localisation. not be used because oxygen may affect the labelling procedure. Related Articles: Angiogram, k-space, Phase encoding Excess pressure can be avoided by withdrawing an equal vol- K Further Reading: Van Vaals, J. J. et al. 1993. ‘Keyhole’ ume of gas with the same syringe. The labelling instructions for method for accelerating imaging of contrast agent uptake. J. Myoview™ (GE Healthcare) however state the use of a breather Magn. Reson. Imaging 3:671–675. needle since nitrogen can influence the radiochemical purity. Some radiopharmaceuticals should also be heated in a boiling Kit water bath in the labelling procedure. The instructions of the (Nuclear Medicine) Kit, or cold kit, is a prefabricated product labelling procedure given by the manufacturer should be fol- containing the nonradioactive chemicals needed to produce a spe- lowed strictly. cific 99Tcm-radiopharmaceutical after adding the required activity Further Readings: Kowalsky, R. J. and S. W. Falen. 2004. 99Tcm-pertechnetate. The chemical substances in the sterile vial Radiopharmaceuticals in Nuclear Pharmacy and Nuclear (kit) consist of the complexing agent (ligand), reducing agent, Medicine, 2nd edn., American Pharmacists Association, for example stannous chloride, stannous fluoride, or stannous Washington, DC; Murray, T., A. A. Bolster, T. E. Hilditch and tartrate, stabilisers, dispersing agents, transfer ligands and buf- A. T. Elliott. 2000. Technetium-99 m-tetrofosmin: Retention fers, all in a nitrogen atmosphere. The solution is lyophilised to of nitrogen atmosphere in kit vial as a cause of poor quality remove all water to extend the stability of the kit. Pertechnetate material. Nucl. Med. Commun. 21:845–849; Saha, G. B. 2004. obtained from the generator does not label to the pharmaceutical Fundamentals of Nuclear Pharmacy, 5th edn., Springer, New without reduction to a lower valency state by the reducing agent. York; Zolle, I., ed. 2007. Technetium-99 m Pharmaceuticals – The stannous ion, however, is easily oxidised and therefore the kit Preparation and Quality Control in Nuclear Medicine, Springer, is preserved in a nitrogen atmosphere. Heidelberg, Germany. Klein–Nishina differential cross section 526 Klystron Klein–Nishina differential cross section 2 1 (Radiation Protection) The Klein–Nishina differential cross sec- tion attempts to describe mathematically the result of interactions F F between incident photons of ionising radiation with single elec- trons in the absorbing medium. It describes the probability that a 3 photon will be scattered to any particular angle, with associated 3 A B 1 transfer of energy to the electron. The probability of scatter to any particular angle, for a given V V C D F energy of incident photon, can be represented graphically thus (Figure K.5). FIGURE K.6 The KLM equivalent circuit model. In the upper part of The Klein–Nishina differential cross section and equation, the figure are the two acoustic ports 1 and 2, connected to the electrical Bethe–Bloch equation, together with the, Molière scattering the- port 3, lower left. To the lower right is an ABCD matrix representation of ory and others, attempts to describe interactions between ionising the model, between port 3 and port 1. radiation and matter at an atomic level, and forms the mathemati- cal basis for radiation dosimetry based on Monte Carlo statistical modelling. the acoustic backing. The load applied here can be used to modify Related Articles: Klein–Nishina equation, Bethe–Bloch equa- the bandwidth and sensitivity. If desired, additional loads can be tion, Molière scattering theory added on port 1 representing matching layers, etc. In the lower right part of the figure is shown how the KLM Klein–Nishina equation model can be collapsed into a single ABCD matrix between the (Radiation Protection) The Klein–Nishina equation describes electrical port and the forward acoustic load. Such a matrix can mathematically the probabilities of interactions occurring when easily be implemented, for instance, in MATLAB®. photonic ionising radiation is incident on an absorbing medium. For more detail, see the article on Klein–Nishina differential Klystron cross section. (Radiotherapy) Klystrons are used in linear accelerators as Related Article: Klein–Nishina differential cross section sources of high-power microwaves, which accelerate the electrons Further Reading: Podgorsak, E. 2010. Radiation Physics for in the waveguide. An alternative option for the microwave source Medical Physicists, Springer, Berlin, Germany. is a magnetron. A klystron is essentially a microwave amplifier that requires KLM equivalent circuit model an initial source of low-power microwaves, and is illustrated in (Ultrasound) The Krimholtz, Leedom, Matthaei (KLM) model Figure K.7. An electron cloud, produced by a cathode filament, is an equivalent circuit model that describes the electro-acoustic accelerates towards the zero potential of an electron beam col- coupling for a piezoelectric transducer. It is a development of the lector via a series of buncher cavities. The first cavity receives Mason model, mainly by the separation of the acoustical and low-power microwaves such that the electric field varies sinusoi- electrical parts of the transmission process. In Figure K.6, the dally across the cavity walls. This causes the electrons to bunch electrical part is in the lower part of the figure (port 3) and the together as they undergo varying acceleration, and move at a fre- acoustical part in the upper part (ports 1 and 2). Port 1 represents quency depending on the resonant frequency of the cavity. The the transmission into water or the body, while port 2 represents electrons then travel through connecting, magnetically steered drift tubes towards the final catcher cavity, which will resonate at the electron arrival frequency, and produce the amplified RF wave. This field can then be transferred to the main accelerating 90 8e–030 2.75 eV waveguide via intermediate waveguides. 120 60 60 keV Klystrons are relatively large compared to magnetrons and 511 keV 1.46 MeV hence normally are mounted behind the gantry. This complicates K 10 MeV 150 4e–030 30 Probe Buncher 180 0 Electrostatic Heater shield Cathode 150 Catcher 30 Drift tube 120 60 – + 90 FIGURE K.5 Distribution of scattered photons at various energies. FIGURE K.7 A typical klystron arrangement. k-space 527 k-space trajectories the transfer of the microwaves from the klystron to the wave- Further Readings: Ljunggren, S. 1983. A simple graphi- guide, which is mounted within the gantry. They also create a cal representation of Fourier-based imaging methods. J. Magn. source of kilovoltage x-rays from the surroundings, absorbing Reson. 54:338; Twieg, D. 1983. The k-trajectory formulation of the accelerating electrons. However, they are more reliable and the NMR imaging process with applications in analysis and syn- hence need replacing less often, which offsets the greater initial thesis of imaging methods. Med. Phys.

10:610. cost. They are also more stable, and do not suffer from sensitiv- ity to the Earth’s magnetic field fluctuations as magnetrons do. k-space trajectories Related Articles: Linac, Magnetron (Magnetic Resonance) The k-space trajectory is the path traced through k-space in a given MRI imaging sequence. Any given Krypton-81m point in 2D k-space has coordinates (kx,ky), representing spatial (Nuclear Medicine) A radionuclide used for in-vivo gamma frequencies in the x and y directions, respectively. Being ‘at a imaging. point in k-space’ means that, at a particular instant, spin phase angles across a slice in an object have a spatial distribution repre- sented by the particular spatial frequencies (kx,ky) at that point in k-space. A signal sample taken at that instant and written into the Half-Life Photon Emission Common Application location in k-space is related to the amplitude of that spatial fre- 13 seconds 190 keV Lung imaging (ventilation scan) quency component (kx,ky) in an image of that slice. By sweeping through k-space and filling it with signal samples, sufficient data is acquired such that a Fourier transform of k-space will recover 81mKr gas is an agent used in lung ventilation imaging, typically an image of a slice through the object. in planar gamma imaging. However, due to the ultrashort half-life The gradient fields control the k-space trajectory in the (kx,ky) its distribution within the lung can only show regional variation. plane. Representing location in k-space as a time-varying vector Also, the 81Rb / 81mKr generator is expensive. Due to these rea- k(t), and the gradient field as a time-varying vector G(t), the tra- sons, the agent is not widely used in routine clinical practice. jectory is given by Related Articles: Gamma camera, Radionuclide imaging Further Readings: Cherry, Sorenson and Phelps. 2012. t Physics in Nuclear Medicine, 4th edn., Elsevier; Mettler and k (t ) g = 2 òG (t ) × dt (K.2) p Guiberteau. 2012. Essentials of Nuclear Medicine Imaging, 6th 0 edn., Elsevier; Zeissman, O’ Malley, Thrall and Fahey. 2014. Nuclear Medicine, 4th edn., Elsevier. where γ is the gyromagnetic ratio. Separating the kx and ky com- ponents of the vector k(t) along the x and y directions in k-space, and defining one gradient direction as phase encode and the k-space orthogonal direction as frequency encode, the trajectory can be (Magnetic Resonance) The k-space is a formalism to describe rewritten in simplified form as gradient encoding of the MRI signal prior to Fourier transform (independently introduced in 1983 by Ljunggren [1983] and t Twieg [1983]). k The position in k-space of each complex data point time is x (t ) g = G t d 2p ò freq ( ) t (K.3) given by the integral of gradients applied between time of excita- 0 tion and acquisition: t g ky (t ) = p òGphase (t )dt (K.4) 2 k = g ò dtG (t ) 0 where γ is the gyromagnetic ratio. The k-space of a slice is two- A Cartesian trajectory is illustrated in Figure K.8, using the dimensional (2D); three dimensions encode a volume. The MR example of a spin-echo sequence. The time course of the tra- image is calculated by a 2D or 3D discrete Fourier transform: jectories as described by (K.3) and (K.4) in the preceding text K can be clearly seen, with the phase encode gradient controlling DFT row to row movements in k-space and the frequency encode S x S ( ( ) Û k ) gradient sweeping the trajectory across columns in k-space. In the Cartesian raster trajectory, all acquired lines in k-space are Thus, the spatial distribution of signal in real space S(x ⃗) corre- equally spaced and parallel. sponds to a distribution in k-space S(k ⃗). S(k ⃗) is usually a com- Spiral filling trajectories may be used in echo-planar imaging plex number (comprising magnitude and phase). (EPI). In a spiral acquisition, central k-space data is acquired at In analogy to optical imaging, the ‘spatial frequencies’ are the start of the scan, allowing for very short effective TE times. called k-vectors and correspond to plane-wave modulations along In a ‘projection acquisition’, each acquired line follows a radial their direction. Most of the image signal is located around the trajectory from the centre of k-space. The data in a radial line k-space origin. High frequencies encode fine detail in the image. through k-space represents the Fourier transform of a projection Thus, the pixel resolution is inversely related to k-space coverage; of the object in a direction orthogonal to the radial line (‘projec- the field-of-view (FOV) to k-space resolution. In standard spin- tion slice theorem’). Acquisition of many radial lines and use of warp phase encoding, k-space and image space have the same a back-projection algorithm can be used for image reconstruction number of rows and columns, unless zero-filling, regridding, in a manner analogous to CT imaging. Alternatively, radial data partial acquisition is applied. The k-space of a magnitude image can be regridding to conventional Cartesian k-space and a Fourier (after zeroing the phase by the modulo-operation) has hermitian transform used to recover the image from k-space. Variants on the symmetry. projection technique were used in the first MR imaging schemes, kV meter 528 kV selector 90° 180° RF transmit ky C TR FID Echo C RF B, D receive TE kx A Gfreq. A B C D Gphase FIGURE K.8 Cartesian k-space trajectory in a spin-echo sequence. Simultaneous applications of the phase encode and frequency encode gradients sweep the trajectory from A to B. The 180° pulse flips the trajectory to point C at the start of a row, and the second application of the frequency encode gradient sweeps the trajectory along a row from C to D. Varying the phase encode amplitude from TR to TR sweeps the trajectory to a different row in k-space at each repetition (e.g. dotted trajectory shown). prior to the now more prevalent spin-warp approach. Radial fill- ing incurs a time penalty of a factor of π relative to Cartesian fill- ing for the same image resolution. In a PROPELLER sequence, the k-space trajectory is a combi- nation of Cartesian and radial type filling. In a single TR, several lines of data parallel to a radial line through the centre k-space are acquired in a fast spin-echo type acquisition. At the next TR, the direction of acquisition is rotated and a new ‘blade’ of data is acquired. The centre of k-space is acquired for each blade and this data redundancy can be used to correct for rigid motion. PROPELLER is useful in the correction of motion artefact in the head. Further Readings: Bock, M. 2002. k-space and resolution. In: Magnetic Resonance Angiography, eds., I. P. Arlart, G. M. Bongartz and G. Marchal, Springer-Verlag, New York; Liang, Z. P. and P. Lauterbaur. 2000. Principles of Magnetic Resonance Imaging: A Signal Processing Perspective, IEEE Press, New FIGURE K.9 Typical non-invasive kVp meter (on the right), with a set of York. different filters that have to cover the two detectors. The filter pack in the lower left corner is upside down, showing the two metal filters. kV meter (Diagnostic Radiology) The high voltage supplied to the x-ray K tube is measured by a kV meter. There are various kV measuring in mammography, CT, etc.) have filters made of materials with devices. specific K absorption edge – for example molybdenum, rhodium, The direct kV measuring device uses a special high-voltage erbium, platinum, etc. divider, which provides low voltage proportional to the high volt- While the non-invasive kV meter seems like a simple device, age and this way allows the use of a low-voltage meter. The insu- the diagnostic physicist should be aware of some of the issues lation between the high-voltage circuit and the measuring device associated with its use. These meters have a specified tolerance, often uses opto-couplers. usually 1–2 kV, so the physicist should not make recommenda- The non-invasive kV device (Figure K.9) usually measures tions for adjustment of the equipment, which are more precise the kV peak (kVp meter). It makes indirect measurement of the than the precision of the meter. kv meters are affected by the angle kVp by assessing the peak energy in the x-ray spectrum. This between the radiation beam and the filter pack, hence the location device has two photo diodes as detectors of the x-ray radiation. and tilt of the kV meter are specified by the manufacturer. Before reaching the detectors, the x-rays beam is attenuated by As with all other equipment, the meter should undergo an special metal filters (different for each detector). This way the two acceptance testing procedure and should be calibrated as recom- detectors produce different signals, depending on the x-ray beam mended by the manufacturer. maximal energy and the attenuation by the filters. Both signals pass through a differential amplifier, which produces a resulting kV selector signal. The latter carries information about the maximal x-ray (Diagnostic Radiology) The kV selector is part of the high-voltage beam energy (kVp). A simpler device would have two copper fil- control circuit of the high-voltage generator (HVG) of an x-ray ters with different thicknesses. More specific kVp meters (used equipment. In classical HVG, the kV selector normally uses one kVp 529 KZK equation or two (broad and fine) switches to select different number of to obtain information about the amplitude of the heart wall move- turns from autotransformer – i.e. different input voltages to the ment. In electro Kymography the exposure, after the narrow beam high-voltage transformer (HVT). A gliding graphite roll can also is passed through patient, is measured with a radiation detec- be used instead of switches (see the diagram in the article High- tor and its electrical signals are compared, again as a source of voltage generator). information about the heart wall movement. Kymography has not Contemporary x-ray equipment with high-frequency generator been used since the 1960s, but its idea has been utilised in other control the kV by varying the frequency of the DC–AC converter. methods. In this case, the kV selector sets the necessary frequency, feed- Further Readings: Chamberlain, E. 1947. Roentgen elek- ing the high-voltage ferrite transformer (see the article on High- trokymography. Acta Radiol. 28:5–6, 847–858. https :/ /ww w .tan frequency generator). dfonl ine .c om /do i /pdf /10 .3 109 /0 0016 9 24709 13802 9; Scott, W. During the quality control procedure, the kV settings are nor- and S. Moore. 1936. Roentgen kymography in diseases of the mally measured indirectly by a kVp meter. A good equipment will heart. JAMA 107(24):1951–1954. have kV with error of the accuracy on the order of 1%. Usually, error above 5% is an indicator of problem in the kV control circuit. KZK equation Related Articles: High-voltage generator, High-frequency (Ultrasound) The KZK equation, or Khokhlov–Zabolotskaya– generator, High-voltage control device Kuznetsov equation, is a wave equation that takes into account non-linear propagation, and also diffraction and non-linearity. kVp The equation can be considered a parabolic (one-way) wave (Diagnostic Radiology) See Peak kilovoltage (kVp) equation including a term describing the losses, and quadratic source term for the non-linear contribution. The equation can be Kymography solved using techniques that can be divided into three catego- (Diagnostic Radiology) Kymography is a historical imaging ries: frequency-domain, time-domain, and combined time- and method from the 1930s – the first method used for visualising mov- frequency-domain methods. The so-called Bergen code is an ing anatomical objects. Its main application has been for recording example, which is available for download on the Internet. The the heart motion. The method uses a relatively long x-ray exposure solutions are computationally intensive, as for each spatial grid (or fluoroscopy) with a thin metal slit (or a number of slits, each point, all intervening beam shapes must be calculated first so about 0.5 to 2 mm), which move in front of the x-ray tube during that the required amount of cumulative waveform distortion can the exposure, thus creating a thin moving x-ray beam. The result develop. Other types of wave equations that describe non-linear of this ‘x-ray scanning of the heart’ is a sequence of lines over the propagation are one-dimensional diffusion (Burger’s), and full- film, each showing the heart wall at different phases. wave equations (Westervelt). In Roentgen Kymography the moving slit creates a ‘jagged Further Reading: Cobbold, R. S. 2007.

Foundations of radiograph’, where the distance between the ‘dents’ is measured Biomedical Ultrasound, Oxford University Press, Oxford, UK. K Volume II L–Z L L Labelling Lamb wave (Nuclear Medicine) Labelling refers to the process of attaching a (Ultrasound) A lamb wave, or plate wave, is an ultrasound wave, radioisotope to a pharmaceutical, also called tracer. The resulting similar to a surface wave propagating in a thin plate with dimen- compound is referred to as a radiopharmaceutical. One example sions comparable with the wavelength. Lamb waves propagate in is the labelling of 18F to FDG (fluorodeoxyglucose). For radio- the same direction as the plate surface. Lamb waves are used in isotopes routinely used in clinical setting, like 99mTc, there exists non-destructive testing of, for example steel plates and wires. a number of labelling kits where each kit is biochemically engi- Related Articles: Longitudinal wave, Transversal wave, neered to target a specific biological process. Surface wave Labelling efficiency Lambert–Beer’s law (Nuclear Medicine) The fraction of radioisotopes successfully (Radiation Protection) The optical radiation passing through labelled to a targeting compound is described by the labelling matter interacts with its molecules due to absorption and scat- efficiency. tering. So the incident radiation fluence Φ (number of photons For many 99mTc pharmaceuticals the labelling efficiency can per unit area) will decrease (Φ − dΦ) after passing the sample of reach >95% if the preparations are performed in accordance with thickness dx (Figure L.2). the manufacturers’ instructions. If the sample is a homogenous solution of the molar concentra- tion c, the decrease of fluence, dΦ, can be expressed as Labelling yield (Nuclear Medicine) The term ‘labelling yield’ is used as a syn- -dF = sFNAc dx onym of labelling efficiency. See Labelling efficiency. where Lactate σ is the cross section (probability) of interaction (Magnetic Resonance) Lactate (lactic acid; Lac) is a chemical N is the Avogadro’s number compound that features in proton (1H) NMR spectra of the brain. The principal resonance seen is a characteristic doublet at 1.32 and after integration the expression ppm due to methyl protons. There is also a quartet at 4.10 ppm F = F - 0e sNcx due to –CH protons, close to the strong water peak (to observe the CH-quartet in in vivo spectroscopy would be a quality criterion [Figure L.1]). This is the so-called Lambert–Beer’s law. It is usually presented It is often difficult to identify the 1.32 ppm resonance unequiv- in the following form: ocally, because contaminating lipid signal appears in the same region of the spectrum. Simple spectral editing can be used to F = F - 010 ecx identify the Lac resonance, since, because of scalar coupling, the doublet is 180° out of phase with the rest of the spectrum in spec- where ɛ is molar decadic extinction coefficient (mol/dm3). tra collected with an echo time of 136 ms and in phase at 272 ms. The Lambert–Beer’s law is used to determine the ionising Alternatively, a double-quantum filter may be used to eliminate radiation dose with Fricke’s dosimeter. It is named after the Swiss also lipid contamination, for example, in skeletal muscle and to physicist Johann Heinrich Lambert (1728–1777) and the German follow the dynamic behaviour of Lac build-up and recovery. physicist August Beer (1825–1863). The concentration of lactate in the normal human brain is very Related Article: Fricke dosimeter low (≈0.5 mM), so the resonance is not visible in healthy subjects. Further Readings: Kiefer, J. 1990. Biological Radiation However, the peak increases in size dramatically in hypoxia/isch- Effects, Springer-Verlag, Berlin, Germany, p. 24; Serdyuk, I. N., N. aemia, reflecting generation of lactate as the end product of anaer- R. Zaccai and J. Zaccai. 2007. Methods in Molecular Biophysics obic metabolism. Thus observation of significant levels of Lac in Structure, Dynamics, Function, Cambridge University Press, the brain is indicative of abnormal energy metabolism due either Cambridge, New York, p. 519. to hypoxia (e.g. within tumours or following ischaemic stroke) or to mitochondrial abnormalities. Laminar flow Related Articles: Editing, Spectral, Magnetic coupling, (Ultrasound; General) Laminar flow describes flow that moves Magnetic resonance spectroscopy in layers. In a tube these layers are parallel to the tube walls. For Further Readings: de Graaf, R. 2007. In Vivo NMR constant flow in a rigid parallel circular tube, the very thin layer Spectroscopy: Principles and Techniques, John Wiley & Sons, of fluid in contact with the wall does not move and there are pro- Chichester, UK; Graham, G. D. et al. 1995. Clinical correlates gressively higher velocities towards the centre of the tube where of proton magnetic resonance spectroscopy findings after acute the velocity is maximum. There is no mixing of flow between cerebral infarction. Stroke 26:225–229. the laminae; if dye is introduced into such flow it can be seen 533 Lanthanides 534 Laser OH are known as lanthanides. The name is sometimes used syn- L –OOC CH CH onymously with the term ‘rare earths’, although the latter 3 includes scandium and yttrium, and excludes the lanthanide promethium. FIGURE L.1 Molecular structure of lactate. Lanthanides are silvery white metals which oxidise when exposed to air. Their f shells contain between 1 and 14 electrons. They have the unusual property that their atomic radius reduces with atomic number due to decreased shielding of the nuclear charge with successive filling of the f shell. Use in Medicine: Solid state lasers: Lanthanides are primar- ily used in solid state lasers, where they provide the impurities Φ Φ – dΦ necessary for light amplification within the gain medium, which can be a crystal, glass or ceramic. A common laser gain medium for medical lasers is the yttrium aluminium garnet (YAG) crystal, which can be doped with small quantities of various lanthanides. Some lanthanide-doped YAG lasers are listed next, along with dx their common medical applications. FIGURE L.2 Attenuation of an optical radiation in the sample of a Nd3+:YAG Ophthalmology, dermatology, dentistry thickness dx. Yb3+:YAG Laser scalpel Er3+:YAG Dentistry, cosmetic surgery, non-invasive blood sugar monitoring Ho3+:YAG Dentistry, orthopaedic surgery, kidney stone and tumour ablation Nuclear medicine imaging radionuclide: Among the lanthanides is gadolinium, whose radioactive isotope 153Gd is widely used in nuclear medicine as a transmission source for quality control and attenuation correction. FIGURE L.3 Flow laminae in a tube. Lanthanum oxybromide in intensifying screen (Diagnostic Radiology) See Rare earth screen Smooth flow profile Blunt flow profile Laplace transform (Nuclear Medicine) The Laplace transform is an integral trans- form used to solve physical problems. The Laplace transform is used to transform an unsolvable differential equation into a solv- able algebraic expression. Velocity profile Velocity vector Hyperlink: http: / /mat hworl d .wol fram. com /L aplac eTran sform .html FIGURE L.4 Smooth and blunt velocity flow profiles. Larmor equation (Magnetic Resonance) Due to the torque exerted by a magnetic to travel in a distinct streamline not mixing with adjacent flow field on a magnetic dipole, it undergoes a precession when ori- (Figure L.3). ented orthogonal to the field. The Larmor equation states the The flow profile describes the shape of the velocity vectors proportionality between the angular frequency of precessing across the tube. With constant steady flow in a tube with constant transverse magnetisation and the flux density B circular cross section, the flow profile becomes parabolic. In this particular case, peak velocity is twice the mean velocity in the w = g B tube. This is not the case in large arteries which have pulsatile flow and tapering vessels, but can be a useful approximation in where some circumstances. Flow is largely laminar in the normal cir- ω is called Larmor frequency culation with disturbed or turbulent flow found in some disease γ is the gyromagnetic ratio states. The flow profiles in arteries with laminar pulsatile flow are described by the Womersley equations (Figure L.4). It appears as part of the Bloch equations. Related Articles: Reynolds number, Turbulent flow Related Articles: Gyromagnetic ratio, Bloch equation Lanthanides Laser (General) The elements in the periodic table that have atomic (Non-Ionising Radiation) The acronym LASER stands for Light numbers between 57 (lanthanum) and 71 (lutetium) inclusive Amplification by Simulated Emission of Radiation, and this Laser 535 Laser describes the process by which a laser beam is formed within the 1. Spontaneous emission – some excited atoms will spon- lasing medium and the laser cavity. taneously lose the energy they have gained and drop L The basic components of a laser are the same for all lasers: back down to ground level, A. When this happens, the an energy source and the lasing material or active medium which atom loses energy and emits a photon with energy (A* may be a solid, liquid or gas. The lasing material is contained − A), a ‘trigger’ photon (Figure L.6b). The value of (A* within the laser cavity which has mirrored ends and acts as a reso- − A) depends on the individual energy levels of par- nant cavity for the amplification of optical radiation. A basic laser ticular lasing materials and can be used to calculate the construction is shown in Figure L.5. frequency of the emitted photon as in Equation L.1. In basic laser theory, the energy source provides energy to the lasing medium, called laser pumping. This energy excites atoms A* - A = hf (L.1) within the lasing material and raises them from ground level (A) to a higher energy level (A*). Following excitation of the atoms where: within the lasing material, the generation of a laser beam depends h = Planck’s constant upon three processes (Figure L.6): and f is the frequency of the photon (Hz) FIGURE L.5 Laser construction. FIGURE L.6 Laser processes in a two-level laser pumping system. Laser aperture 536 Laser beam This can in turn be used to calculate the wavelength of Related Articles: Laser pumping methods, Laser beam deliv- L the photon by applying Equation L.2. ery system c = f l (L.2) Laser aperture (Non-Ionising Radiation) The laser or optical device aperture is where: the exit window of the device, through which the laser or optical c is the speed of light = 3 × 108 m/s radiation beam passes out of the laser unit. The aperture of a laser λ is the wavelength of the photon (m) device must be labelled appropriately (Figure L.7) and in accor- 2. Stimulated emission – if a trigger photon encounters an dance with BS EN 60825-1. excited atom it will force the excited atom to emit a pho- Further Reading: BS EN 60825-1:2014. Safety of laser prod- ton that is identical, i.e. with the same wavelength and ucts: Equipment classification and requirements. phase. The light photons emitted are reflected by the mirrors at either end of the laser cavity. They continue Laser beam to pass through the lasing material and cause further (Non-Ionising Radiation) Laser beams deliver energy in the form stimulated emission and amplification of the original of electromagnetic photons. The behaviour of the laser beam is trigger photons (Figures L.6c and d). governed by the electromagnetic spectrum and associated for- 3. Population inversion – the majority of the lasing mate- mulae. In general, a laser beam has the following characteristics rial atoms are excited rather than in the ground state. and it is these that set the laser apart from other sources of non- Therefore, there is an increased probability that a trig- coherent optical radiation: ger photon will encounter an excited atom and amplifi- cation is maximised (Figure L.6a). • Monochromatic – the beam is made up of photons with a single or small number of discrete wavelengths. When sufficient light photons have been generated to produce a • Low divergence – the photons in the beam spread out laser beam with a certain power, the photons are released via the from the axis of the main beam very slowly and so the partially reflecting mirror and a beam shutter. The beam is deliv- beam maintains its power or energy per unit area char- ered to its point of application via a beam delivery system. acteristics over long distances. • Coherence – all photons in the laser beam have the same frequency and wavelength and are in phase with each other. The wavelength of the laser beam and how that wavelength inter- acts with the body determines how the laser will be used prac- tically in medicine. Where the

absorption coefficient for a laser wavelength is high the laser will be most effective. For example, haemoglobin has an absorption peak at 532 nm (green), therefore greenlight lasers with a wavelength of 532 nm are well absorbed in, and effective at treating, tissue with high haemoglobin content FIGURE L.7 British Standard laser aperture label. (Figure L.8). FIGURE L.8 Absorption and penetration depth in water and other biological tissue constituents for different wavelengths. Laser beam delivery system 537 Laser film printer TABLE L.1 L Laser Beam Properties Power (W) Rate of energy delivery, 1W = 1J/s Laser spot size (cm2) The area of the laser beam spot size at the treatment surface. Laser beam profile The uniformity of the laser beam over the exposure area Irradiance (W/cm2) or power density Power delivered per unit area. Laser exposure duration (s) (pulse width for pulsed lasers) Time over which energy is delivered. Energy per pulse (J) For a pulsed laser beam, energy per pulse (J) = power per pulse (W) x pulse duration (s) Pulse duration (s) Length of time a laser pulse lasts Fluence (J/cm2), also dose, radiant exposure or energy Amount of energy delivered per unit area density Pulse repetition rate or pulse frequency (Hz) Number of laser pulses per second Pulse profile Shape of the laser pulse with time Energy (J) Energy delivered by a laser treatment is the sum of all the individual pulse energies delivered during the treatment. Laser beams are described using the parameters shown in risk, the nature of the laser radiation, i.e. visible or invisible laser Table L.1. All of these parameters will have an effect on laser Class, the laser wavelength and the name and publication date of treatment. the standard to which the laser has been classified. Related Articles: Photons, Electromagnetic spectrum, Laser Related Articles: Laser aperture, Accessible emission limit output mode, Irradiance (AEL), Nominal ocular hazard distance (NOHD) Further Readings: Nouri, K. 2011. Lasers in Dermatology Further Readings: BS EN 60825-1:2014. Safety of laser and Medicine, Springer-Verlag, ISBN 978-0-85729-280-3; products: Equipment classification and requirements; Council Scholle, K., S. Lamrini, P. Koopman and P. Fuhrberg. 2010. 2 µm Directive 2006/25/EC on the minimum health and safety require- laser sources and their possible applications. Frontiers in Guided ments regarding the exposure of workers to risks arising from Wave Optics and Optoelectronics, IntechOpen, ISBN 978-953- physical agents (artificial optical radiation) (19th individual 7619-82-4, www .i ntech open. com /b ooks/ front iers- in -gu ided- Directive within the meaning of Article 16(1) of Directive 89/391/ wave- optic s -and -opto elect ronic s. EEC) [2006] OJ L 114; A Non-Binding Guide to the Artificial Hyperlinks: www .i ntech open. com /b ooks/ front iers- in -gu ided- Optical Radiation Directive 2006/25/EC, Radiation Protection wave- optic s -and -opto elect ronic s; www .i ntech open. com /b ooks/ Division, Health Protection Agency. front iers- in -gu ided- wave- optic s -and -opto elect ronic s /2 -m -lase r -sou rces- and -t heir- possi ble -a pplic ation s Laser film printer (Diagnostic Radiology) The laser film printer (also known as Laser beam delivery system dry laser imager or laser camera) produces a copy of a digital (Non-Ionising Radiation) Laser beam delivery systems are used image on a special type of film (often transparent film, similar to to deliver the laser beam from the laser aperture to the treatment x-ray film). The laser film printer is used to produce hard copies area. Some common types of system, and the applications they of images from various digital imaging systems, as CT scanners, are often used for, are shown in Table L.2. digital radiography, MR, ultrasound, gamma camera, etc. Hyperlinks: www .t opcon medic al .co m /pro ducts /pasc alstr The laser film printer uses direct laser scanning (projecting the eamli ne -pu blica tions .htm; www .g yneas .com/ colpo scope -opti que image) onto the film. The device often uses infrared solid state -s eiler -985- led .h tml; www .l umeni s .com /medi cal /c o2 -pr oduct s / laser with very small laser spot spacing (e.g. 50 μm). The initial ult rapul se -du o/; www .thorlaser .com /pet/ laser source intensity is constant, but before reaching the film this intensity is changed by a laser modulator to produce intensities, Laser classification responding to the grey levels of the pixels in the digital image. (Non-Ionising Radiation) Compliance with the relevant parts The modulated laser beam scans the film and exposes it to pro- of BS EN 60825 enables laser products to meet applicable legal duce an accurate film copy of the digital image. After the expo- requirements. Detailed within the standard is the recommended sure the film is developed. process of laser classification. The film for a laser film printer usually contains cubic grains The eight laser classes and their associated hazards are and not tabular grains (as in x-ray screen-film); however, the described in Table L.3. In accordance with EU Directive 2006/25/ printer can also work with normal x-ray film and wet-type film EU, all Class 3B and Class 4 lasers in use in the workplace must processing. Variations in image quality resulting from less than be adequately risk assessed. optimal wet film development are potential problems. A newly It is a recommendation of BS EN 60825-1 that lasers be developed thermographic film developer for laser films (without labelled with a laser trefoil, that the laser aperture be labelled liquid powdered chemicals) is environmentally better and reduces and that information depending on the class of laser is given. It operating costs. is recommended that Class 3B and Class 4 lasers are labelled as X-ray film with dry processing methods (used with some laser shown in Figure L.9 with a warning of the hazards, the organs at film printers): Laser film printer 538 Laser film printer L TABLE L.2 Laser Beam Delivery Devices and Examples of Their Applications Delivery Device Applications Optical fibre – very small diameter pipe made out of glass or plastic, used Widespread use across all laser to transmit light. applications. Optical fibre and slit lamp Ophthalmology www .t opcon medic al .co m /pro ducts /pasc alstr eamli ne -pu blica tions .htm Optical fibre and endoscope Many internal laser applications, e.g. urology, gynaecology, ear, nose and throat (ENT) Optical fibre and colposcope Gynaecology www .g yneas .com/ colpo scope -opti que -s eiler -985- led .h tml Optical Fibre and Handpiece Physiotherapy, hair removal, skin resurfacing, tattoo removal www .thorlaser .com /pet/ Laser film printer 539 Laser film printer TABLE L.2 (CONTINUED) L Laser Beam Delivery Devices and Examples of Their Applications Delivery Device Applications Articulated arm Incorporates a mirror system to transmit the beam. Used with carbon dioxide lasers as the 10,600 nm wavelength is absorbed when travelling through most optical fibres. www .l umeni s .com /medi cal /c o2 -pr oduct s /ult rapul se -du o/ TABLE L.3 Laser Classes, Description of Associated Hazards and Example Devices Class Description of Associated Hazard Example Device Class 1 Safe under conditions of normal use. Laser in a CD or DVD player Class 1M Safe under conditions of normal use, may be hazardous when using viewing optics. Class 1C Designed for application to the skin or non-ocular tissue. Irradiant or radiant exposure levels may Home hair removal device exceed the skin maximum permissible exposure (MPE) as necessary for intended treatment. Ocular hazard is prevented by engineering means during use, i.e. AEL is below Class 1 when the applicator is not in contact with the skin/non-ocular tissue. www .g ov .uk /gove rnmen t /pub licat ions/ laser -radi ation -safe ty -ad vice/ laser -radi ation -safe ty -ad vice Class 2 Safe for short exposures – eye is protected by natural aversion response Bar code scanner Class 2M Safe for short exposures – may be hazardous when using viewing optics. Class 3R Low risk of injury for normal use, may be dangerous if used improperly by untrained individuals. High power laser pointers Class 3B Direct viewing is hazardous. Physiotherapy lasers for LLLT Class 4 Risk of injury to eye and skin. Associated fire hazard. Surgical lasers Abbreviations: M – magnifying optical viewing instruments R – reduced/relaxed requirements for manufacturer & user, e.g. no key switch B – historical C -– contact 1. Adherographic: The film has a laser-sensitive adhesive requires very thin laser and small pells (5 μm): 16 × 5 layer plus imaging layer (carbon particles, pells), both = 80 μm pixel, which will produce image resolution of sandwiched between two polyester sheets. When the 6.25 lp/mm. laser beam scans the dry-film it causes the adhesive 2. Thermal: The film emulsion is a combination of sil- layer to take carbon and stick it to the polyester sheet. ver behenate and silver halide coated onto polyester. As a result there are two sheets with positive and nega- The scanning laser beam triggers ‘thermal developing tive image. The first is coated and used as film, the other process’ producing a ‘true’ greyscale. However there is disposed. The adhesion process is binary and the grey is no fixer – that is, the undeveloped silver halide tone (nuance) is produced by dithering. Normally a cell crystals remain on the film, which makes it thermally of 16 × 16 pells makes a pixel with 256 grey levels. This unstable. Laser interferometry 540 Laser output mode L FIGURE L.9 Examples of Class 3B and Class 4 and laser labels. These laser film printers could produce an image with less grey Photodetector Watertank levels (i.e. less contrast), compared with normal x-ray film. However, this is not necessarily a specific disadvantage, because Laser Beam splitter the digital image comes from a digital imaging system, where the ‘window’ technique has to be used to obtain optimum contrast by Transducer the operator. The throughput of a printer with wet processing is high Membrane (usually more than 100 films per hour). For films with thermal/ Mirror adherographic dry processing the time for producing one film is much longer (often 30–45 s). The image quality depends on FIGURE L.10 Laser interferometry. The beam splitter splits the laser the laser spot and is selectable (e.g. producing 50 or 100 μm beam into two parts. The reference beam is reflected on the mirror and pixel size with 12 bit greyscale resolution). The printer also the other part is reflected on the membrane, which is moving in unison has automatic density control and image interpolation technol- with the ultrasound field. The varying phase difference between the two ogy. The laser film printer (laser imager) allows PACS connec- beams is detected by the photodetector. (Courtesy of EMIT project, www .emerald2 .eu) tion, which makes it very useful for large hospital information systems. Laser localisers Laser interferometry (Radiotherapy) In external beam radiotherapy treatment it is (Ultrasound) A laser interferometry system is used for accurate important to relate the internal treatment target to the delivery high resolution (temporal and spatial) displacement measure- system. A key component of this is to mark the patient’s skin with ments. At the National Physical Laboratory NPL, UK, such a tattoos that are aligned with the delivery coordinate system using system is used as a primary standard instrument for calibrating the treatment room lasers. These are known as skin reference hydrophones. The system includes a laser, a photodetector, mir- marks. Often two lateral and one anterior tattoos and lasers are ror, beamsplitter, thin membrane and a water tank, Figure L.10. used. The equipment is mounted on a table equipped with vibration The first stage of the treatment process involves imaging the damping. patient, for example with CT. The tattoos are often drawn at this A thin acoustically transparent membrane will move in uni- point. At treatment simulation and each treatment visit, localisa- son with the surrounding medium. When placed in an ultrasound tion lasers in each room are used to set the patient up. Often the field it will therefore follow the particle displacement. This set-up is further refined using imaging. movement can be calculated from the measurements of the phase Abbreviation: CT = Computed tomography difference between the fixed optical reference beam (reflected on Related Articles: Skin reference marks, Treatment verification the mirror) and the beam, which has been reflected on

the mov- ing membrane. When the amplitude of the particle displacement Laser output mode s0 is known the pressure amplitude p can be calculated using the (Non-Ionising Radiation) relationship p = Z ωs0. Continuous Mode: The simplest mode in which a laser may Related Articles: Displacement, Hydrophone. operate is continuous mode, that is, the laser beam is continuously Laser plasma particle accelerator 541 Laser plasma particle accelerator emitted and applied via the delivery device. The time for which level is released to give a high energy ‘giant pulse’ with a duration the beam is applied is therefore determined by the operator using in the order of nanoseconds (ns). Q-switching is used to photome- L the exposure switch. The power (W) of continuous beam laser chanically break up material such as kidney stones or tattoo dye outputs is measured to indicate the energy deposited by the beam pigments. per second. When the area over which the beam acts is taken into Related Articles: Laser beam account the irradiance (W/m2), or power deposited per unit area Further Readings: Bertolotti, M. 1983. Masers and Lasers, of the beam, can also be calculated. This is used to indicate the An Historical Approach, Adam Hilger Ltd, Bristol, UK, ISBN potential a particular laser beam and delivery device has to treat 0-85274-536-2; Brown, R. 1968. Lasers, Tools of Modern or cause damage. Technology, Aldus Books, London, UK; Carruth, J. A. S. and Pulsed Mode: Lasers are commonly operated in pulsed mode A. L. McKenzie. 1986. Medical Lasers Science and Clinical and indeed it is often possible to use either continuous or pulsed Practice, Adam Hilger Ltd, Bristol, UK; Henderson, A. R. 1997. mode for the same laser. In pulsed mode the laser has a defined A Guide to Laser Safety, Chapmann & Hall. pulse length and therefore pulse frequency (number of pulses per second). Pulse lengths seen in medical applications are generally Laser plasma particle accelerator in the order of milliseconds (ms) with pulse repetition frequencies (Radiotherapy) Laser plasma particle accelerators (LPPAs) accel- in the order of kilohertz (KHz). erate particles by shining an ultra-high intensity laser beam at As the pulse length of a particular laser pulse is known when a target foil. They are yet to make it to the radiotherapy clinic, using pulsed mode, the output of pulsed laser beams is measured but are of long-term interest to the particle therapy community in Joules (J). Therefore, the power deposited in a specific time is as they have the potential to be highly space-efficient (table-top indicated. When the area over which the beam is acting is taken sized) and lightweight. into account the radiant exposure of the beam (J/m2) may be cal- The most well-understood laser plasma particle acceleration pro- culated to indicate the treatment or damage potential of the beam. cess is called the target normal sheath acceleration (TNSA) (Karsch Q-Switched Mode: Q-switched mode is a specialist laser et al., 2017). A thin target foil is irradiated by an ultra-intense laser mode used in certain applications. A Q-switch is an electro- beam of magnitude ~1018 W/cm2 for the main pulse and ~1012 W/ optical component which stops optical radiation when activated cm2 for the prepulse. The prepulse generates a plasma of electrons and transmits optical radiation when deactivated. Such a device on the target’s front side. These electrons are then accelerated in is placed between the mirrors at either end of the laser cavity the forward direction by the electric field created by the laser main (Figures L.11 and L.12). pulse. The electrons pass through the target foil, creating a strong Initially, while the lasing material is pumped the Q-switch is local charge imbalance. This charge imbalance creates a sheath field activated and lasing cannot occur as the optical path through the at the target rear surface, of magnitude several orders higher than laser cavity is blocked. Pumping energy is stored in the upper the electric field in conventional accelerators. This electric field then energy level, allowing for maximum population inversion. When accelerates the ions from the back of the target foil towards the for- the Q-switch is deactivated the energy stored in the upper energy ward direction, resulting in an expanding ion beam. A few challenges of applying LPPAs in particle therapy are: 1. The ultra-high intensity laser prepulse can damage the target before the main pulse arrives. 2. The output ion beam can consist of different ions origi- nated from target coatings or contamination of the tar- get rear surface. 3. The output ion beam is divergent. Related Articles: Cyclotron, Synchrotron, Synchrocyclotron, FIGURE L.11 Q-switching. Fixed field alternating gradient (FFAG) accelerator FIGURE L.12 Illustration of pulse lengths for different laser modes. Laser protection adviser (LPA) 542 Laser protective eyewear Further Readings: Daido, H., M. Nishiuchi and A. S. Pirozhkov. safety. It is recommended that the LPS or LSO should be suitably L 2012. Review of laser-driven ion sources and their applications. trained and formally appointed. Rep. Prog. Phys. 75(5):056401; Fuchs, J., P. Audebert, P. Antici, The responsibilities of the LPS or LSO will be agreed locally E. Brambrink, E. d’Humières, J. C. Gauthier, L. Romagnani, by all parties and should include: M. Borghesi, C. A. Cecchetti, E. Lefebvre and DPTA, C. 2006. Review of high-brightness proton & ion acceleration using pulsed • Supervision of laser and/or optical radiation work lasers. In Prepared for (pp. 319–323); Hofmann, K. M., U. Masood, • Supervision of laser and/or optical radiation equipment J. Pawelke and J. J. Wilkens. 2015. A treatment planning study • Ensure that the local rules are observed and followed on to assess the feasibility of laser‐driven proton therapy using a a day-to-day basis compact gantry design. Med. Phys. 42(9):5120–5129; Karsch, L., E. Beyreuther, W. Enghardt, M. Gotz, U. Masood, U. Schramm, More information regarding the role of the LPS is available in the K. Zeil and J. Pawelke. 2017. Towards ion beam therapy based MHRA laser guidance document. on laser plasma accelerators. ActaOncologica 56(11):1359–1366; Related Articles: Laser protection adviser (LPA) Ledingham, K., P. Bolton, N. Shikazono and C. M. Ma. 2014. Further Reading: Medicines and Healthcare products Towards laser driven hadron cancer radiotherapy: A review of Regulatory Agency, Lasers, intense light source systems and progress. Appl. Sci. 4(3):402–443; Murakami, M., Y. Hishikawa, LEDs – guidance for safe use in medical, surgical, dental and S. Miyajima, Y. Okazaki, K. L. Sutherland, M. Abe, S. V. Bulanov, aesthetic practices, Crown copyright, September 2015. H. Daido, T. Z. Esirkepov, J. Koga and M. Yamagiwa. 2008, June. Hyperlinks: https :/ /as sets. publi shing .serv ice .g ov .uk /gove Radiotherapy using a laser proton accelerator. In AIP conference rnmen t /upl oads/ syste m /upl oads/ attac hment _data /file /4741 36 /La proceedings (Vol. 1024, No. 1, pp. 275–300). AIP; Roth, M. and se r _g uidan ce _Oc t _201 5 .pdf M. Schollmeier. 2017. Ion acceleration-target normal sheath accel- eration. arXiv preprint arXiv:1705.10569. Laser protective eyewear (Non-Ionising Radiation) Personal protective equipment should Laser protection adviser (LPA) be the last resort with regard to reducing the risk associated with (Non-Ionising Radiation) A laser protection adviser (LPA) may laser use to as low as reasonably practicable. A risk assessment be internal or external to an organisation and has expertise in mat- should be carried out to demonstrate that all possible engineering ters of lasers and laser safety, therefore they are able to assist the and administrative controls have been applied before the use of employer in fulfilling these requirements. There are no defined personal protective equipment has been considered. criteria for LPA competency, further guidance regarding LPA Laser safety glasses and eye shields are the most com- competency may be found via the following sources: monly used laser personal protective equipment in the health- care environment, and they are used by patients and by staff. • Medicines and Healthcare products Regulatory Agency There are many different types of eye protection available. It (MHRA), Lasers, intense light source systems and is important to consider that eye protection is very laser spe- LEDs – guidance for safe use in medical, surgical, den- cific. Different coloured lenses provide protection at different tal and aesthetic practices wavelengths. • Technical report PD CLC/TR 50448 Guide to levels of Laser safety glasses used in healthcare are manufactured competence required in laser safety and tested according to BS EN 207:2009 ‘Personal eye pro- • RPA 2000 LPA certification scheme for laser protection tection equipment. Filters and eye-protectors against laser advisers radiation (laser eye protection)’. Laser glasses used in laser adjustment work must be manufactured and tested according to The MHRA laser guidance document recommends that it is good BS EN 208:2009 ‘Personal eye-protection. Eye protectors for practice for all NHS and private laser providers to consult an LPA adjustment work on lasers and laser systems (laser adjustment with regard to Class 4 and Class 3B laser systems. For private pro- eye-protectors)’. viders, the requirements of the local authority determine whether The optical density of the laser eyewear lenses reduces the the appointment of an LPA is required. The appointed LPA over- incident beam to a beam that is safe for viewing after it has passed sees laser safety within all departments of an organisation. through the lenses of the eyewear. Different pairs of eyewear have Related Articles: Laser protection supervisor (LPS) different optical densities for different laser wavelengths and for Further Readings: Medicines and Healthcare products lasers operating in different modes. The lowest value of optical Regulatory Agency, Lasers, intense light source systems and density is 0, which will give 100% transmission; for every unit LEDs – guidance for safe use in medical, surgical, dental and increase in optical density, transmittance decreases by a factor of aesthetic practices, Crown copyright, September 2015; PD CLC/ 10, as shown in Table L.4. TR 50488:2005 ‘Guide to levels of competence required in laser In addition to this, frames and lenses are tested to ensure safety’ ISBN:0580467309. BSI 2005; RPA 2000: Competence for that they can withstand the beam itself, this is indicated by the Radiation Protection Professionals. scale number (LB) of the glasses. Scale number increases with Hyperlinks: https :/ /as sets. publi shing .serv ice .g ov .uk /gove increased beam attenuation provided. Every unit increase in scale rnmen t /upl oads/ syste m /upl oads/ attac hment _data /file /4741 36 /La number provides a factor of 10 increase in attenuation, e.g. LB2 ser _g uidan ce _Oc t _201 5 .pdf ; www .r pa200 0 .org .uk /l pa -ce rtifi glasses provide ten times the attenuation of LB1 glasses. catio n -sch eme/ BS EN 207 specifies four laser modes as shown in Table L.5. Figure L.13 shows the markings seen on laser safety glasses; Laser protection supervisor (LPS) the laser protection they provide is indicated as follows: (Non-Ionising Radiation) The appointment of a laser protection e.g. if we look at the extract: supervisor (LPS) or laser safety officer (LSO) is recommended at departmental level to assist the LPA in local matters of laser 190–532 nm OD > 9 > 315–532 DM L6 + IR L5 Laser pumping methods 543 Latent image Further Readings: BS EN 207:2009. Personal eye-protection TABLE L.4 equipment. Filters and eye-protectors against laser radiation L Optical Density and Transmission (laser eye-protectors); BS EN 208:2009. Personal eye-protection. Eye protectors for adjustment work on lasers and laser systems Optical Density (OD) Transmission (%) (laser adjustment eye-protectors). 0 100 1 10 Laser pumping methods 2 1 (Non-Ionising Radiation) The energy to achieve population inver- 3 0.1 sion within the laser cavity is provided by the laser energy pump 4 0.01 in a process referred to as ‘pumping’. Different lasing materials 5 0.001 are pumped in different ways, gas lasers tend to use electrical cur- rent, while lasers with a solid state or liquid lasing material tend to etc be optically pumped, for example using a flashlamp. The method used to pump a laser helps to define the properties of that laser. The length of time before spontaneous emission occurs for an excited lasing material atom gives an indication of how easy it is to achieve population inversion in that particular material In TABLE L.5 the two-level pumping system described in Figure L.14, popula- Laser Eyewear Markings tion inversion cannot be achieved as spontaneous emission occurs

very quickly, resulting in a constant transfer of atoms between the Mode Letter Pulse length ground and excited energy levels. Continuous wave D > 0.25 s For some materials it is possible to make use of an energy Pulsed mode I > 1 µs–0.25 s level between the ground and excited states. This allows the Giant pulsed mode R > 1 ns–1 µs atoms in the material to remain in a higher energy state for a longer period of time so that population inversion is more easily Modelocked (ultrashort pulses) M < 1 ns achieved. Where possible, two energy levels between the excited and ground state may be used for a four-level pumping system. This system further prolongs the time for which lasing material atoms are in an excited state. This arrangement is used in neo- dymium doped ytrium aluminium garnet (Nd:YAG) lasers. Related Articles: Laser Lasing material (Non-Ionising Radiation) See Laser Late radiation toxicity (Radiotherapy) See Late response of normal tissue (Radiotherapy) Late reactions (Radiotherapy) See Late response of normal tissue (Radiotherapy) FIGURE L.13 BS EN Laser safety glasses markings. Late response of normal tissue (Radiotherapy) Radiation treatment inevitably affects nor- mal tissue and so may cause radiation-induced adverse effects. • 190–532 and >315–532 indicate wavelength ranges These effects are usually divided into early and late reactions. • OD > 9 indicates that the glasses have an optical den- Traditionally, reactions that occur more than 90 days after the start sity of 9 of treatment are classified as late effects. The type and severity of • D indicates continuous mode late effects depends upon: the body region treated, the radiation • M indicates modelocked mode dose, the radiation technique and the individual radiosensitivity • I indicates pulsed mode of the patient. Late reactions are often long-term or permanent. • R indicates giant pulsed mode Related Articles: Adverse effects (Radiotherapy), Long-term • L(B)6 and L(B)5 indicate the scale number morbidity, Tolerance, Probability of complications, Sigmoid dose-response curve, Dose response model So, between 190 and 532 nm these glasses have an optical density of 9. Between 315 and 532 nm, they have a scale number of L(B)6 Latent image in continuous wave and modelocked modes and a scale number of (Diagnostic Radiology) Latent image (hidden image) is a term L(B)5 in pulsed and giant pulsed mode. used in photography to describe the invisible changes in the Laser glasses are usually not transferrable between different exposed film before its development. These changes are in fact lasers and even the upgrade of an existing laser may mean that clusters of metallic silver atoms formed on the silver halide crys- laser glasses of a different optical density and scale number are tals after exposure to light, as it is well known the size of the required. The LPA should always be consulted with regard to the cluster is proportional to the intensity of the light, thus forming an purchase and use of laser safety glasses. invisible pattern – latent image. The ‘latent image’ term has found Latent period 544 Lateral electronic equilibrium L FIGURE L.14 Laser pumping systems. a stable place in x-ray radiography and other image formation 0 1 2 3 4 5 6 Depth in lucite (cm) mechanisms. For example, the pattern of absorbed x-rays, formed 100 φ = 38 φ = 19 after they have passed through the object/patient (before expos- φ = 13 ing the film/detector) is also called ‘latent image’. Sometimes this φ = 9 φ = 7 latent image is also called modulated (by the object) x-ray beam. 50 φ = 5 Related Article: X-rays, X-ray film φ = 3 Latent period (Radiation Protection) The time interval between a person being exposed to a carcinogen, whether it be a chemical pollutant, ionis- 20 ing radiation or any other factor, and the progression of the dis- ease to diagnosis as a leukaemia or solid cancer is known as the FIGURE L.15 Buildup curves for different photon beam diameter. latent period. This period can extend to tens of years, and may (From Dutreix, J. et al., Phys. Med. Biol., 10, 177, 1965.) make it difficult or impossible to make an association between a particular exposure event and the resulting cancer when car- rying out multi-factorial analyses or epidemiological studies on Lateral electronic equilibrium populations. (Radiotherapy) The condition of transient electronic equilibrium Related Articles: Radiobiological models, Stochastic effects implies that the distance between the point where the dose mea- surement is performed and the limit or edge of the beam is larger Lateral than the maximum range of secondary electrons set into motion (General) Directional anatomical terms describe the relationship by photons. In a photon beam the transient electronic equilibrium of structures relative to other structures or locations in the body. is assured on the beam axis only for broad beams. Near the edge Lateral: Towards the side, away from the mid-line (e.g. the of a beam there is a lack of lateral electronic equilibrium thereby little toe is located at the lateral side of the foot). making accurate dosimetry very difficult. In Figure L.15 build up Related Article: Anatomical relationships curves measured in a polymethylmethacrylate (PMMA) phantom at different depths for a 20 MVXR beam are shown. The diameter Lateral dose falloff of the photon beam ranges from 3 to 38 mm. The lack of lateral (Radiotherapy) The lateral dose falloff is the high gradient region electronic equilibrium is evident in beams whose width is less at each side of a radiation field profile. It is quantitatively char- than 38 mm. The dose within the field decreases with depth, the acterised by the lateral penumbra. In photon therapy, the lateral penumbra is broadened and the dose deposition outside the field dose falloff is influenced by: the photon energy, the initial source edge increases. size, the collimator and the depth of the field. In proton therapy, Dose in small radiation beams is therefore inherently difficult due to uncertainties in range and biology at the distal dose falloff, to measure. Clinical examples where regions of electronic non- the lateral falloff is often used to spare critical organs. It is influ- equilibrium are encountered are intensity modulated radiation enced by: the initial source size; the collimator; scattering in both therapy (IMRT) where small individual beams (beamlets) of less the beamline and in the patient, and in the case of pencil beam than 1 cm2 are commonly employed to create a predefined dose scanning, the optimisation technique. distribution and functional stereotactic radiosurgery (SRS) where Related Articles: Penumbra, Lateral penumbra similarly sized beamlets may be employed. Further Reading: Paganetti, H., ed., 2012. Proton Therapy Related Articles: Electronic equilibrium, Charged particle Physics. CRC Press (pp. 113–117). disequilibrium Percentage depth dose (%) Lateral penumbra 545 Lateral resolution Further Reading: Dutreix, J., A. Dutreix and M. Tubiana. 1965. Electronic equilibrium and transition stages. Phys. Med. L Biol. 10:177–189. Lateral penumbra (Radiotherapy) The lateral penumbra is a quantitative measure for the lateral dose falloff. It is defined using two points along the lateral dose falloff, for example the 80% to 20% dose positions (Figure L.16). Related Articles: Penumbra, Lateral falloff. Resulting image Scan lines Lateral position (General) There is a series of terms used to describe the position of an individual when undertaking different imaging examination. A Lateral: Standing, sitting or lying down with onside in con- tact with the equipment couch or stand. For example, erect lateral chest x-ray. Related Article: Patient position Lateral resolution B (Ultrasound) The lateral resolution of an ultrasound system deter- (a) (b) mines its ability to separate objects in the transverse/lateral direc- tion of the image. It is usually defined as the smallest separation FIGURE L.17 Lateral resolution. (a) Transducer and ultrasound beams of identical point targets at the same depth in the image plane. with pairs of targets A and B. (b) Resulting image of targets. Lateral resolution is dependent on beam width. Beam width itself varies throughout the image and is dependent on the depth, aperture (link) and the focal point of the image. In areas of nar- row width, separation will be better than zones where the beam is wide (Figure L.17). In practice the intensity of the beam is highest in the centre and falls towards the edge of the beam so that the resulting image of the point target is complex and the lateral reso- lution will also depend on power, gain and dynamic range chosen. Figure L.17 shows how lateral resolution is dependent on beam width. In Figure L.17a, three beams sweep across two pairs of targets. At the focal point depth (A), the beam is narrow and the middle beam does not insonate either target. There is a gap seen in the displayed image (Figure L.17b). In the deeper pair (B), the targets are each insonated by two beams and the echoes merge into one another. At the unfocused depth, echoes from the target are always combined with echoes from adjacent tissue with the result that contrast between target and tissue is reduced. (a) (b) FIGURE L.18 Ultrasound images of phantom illustrating focus effect on lateral resolution. (a) Focus superficial to circled pins. (b) Focus at circled pins. In most scanners, some control of the lateral resolution is pos- sible by selecting an appropriate focal depth for the target under examination. This optimises the transmitted beam width for a particular depth where lateral resolution will be optimised. The effect is shown in Figure L.18. More recently, new beam-forming techniques have dispensed with user-focusing. Figure L.18 shows that by altering focal depth (arrow cur- sor to side of image), lateral resolution is optimised to suit the depth of the target under investigation, in this case wires in an ultrasound phantom. In Figure L.18b, the pins at depth (circled) FIGURE L.16 Lateral penumbra. appear round. When the focus is moved more superficially Latitude of film 546 LCD (liquid crystal display) L (a) (b) FIGURE L.19 Ultrasound images of phantom illustrating the effect of the age of the ultrasound machine. (a) Older ultrasound scanner. (b) Newer ultrasound scanner. (Figure L.18a), the deep targets appear wider, and contrast and resolution is reduced. 3.0 Latitude Lateral resolution may be assessed by examining the spread of point targets in a phantom or by examining the separation of groups of point targets. The images in Figure L.19 show the 2.5 improvements in lateral resolution demonstrated by a group of Under Over targets in a phantom. exposure exposure Figure L.19 shows ultrasound images from a phantom with 2.0 nine wire targets, five of which lie in a horizontal line with spac- ing of 2, 1, 0.5 and 0.25 mm. In the older scanner (Figure L.19a) 1.0 only the largest (2 mm) gap is evident, with the other four targets 0.9 merged into one image. In the newer scanner (Figure L.19b), the 1.5 0.8 wires show improved spatial resolution. There are gaps at 1 mm 0.7 lateral spacing and some separation of the echoes from 0.5 mm. Contrast 0.6 The last two pins appear merged as one. 1.0 0.5 0.4 Latitude of film 0.3 (Diagnostic Radiology) The latitude of a radiographic film is the 0.5 0.2 exposure range over which it can record contrast as illustrated in Figure L.20. 0.1 The latitude is a design characteristic of film and is taken into 1 1 1 1 1 1 1 2 4 8 16 32 64 account when selecting film for specific clinical applications. For 64 32 16 8 4 2 example, chest radiography is usually done with a film with rela- Relative exposure tively wide latitude. Related Article: Characteristic curve FIGURE L.20 Latitude of film and contrast. (Courtesy of Sprawls Foundation, www .sprawls .org) Lattice (General) A crystal is a solid form of matter which has transla- tional periodicity in three-dimensions in its atomic arrangement. Crystal structure Crystal structures are described in terms of a lattice plus a motif. Lattice Motif The lattice is an infinite array of points with identical environ- ments and the motif is the element of structure associated with Lattice each lattice point. A vector that joins two lattice points is known vectors as a lattice vector (Figure L.21). The term lattice is also used in magnetic resonance imag- Lattice ing (MRI) to refer to the magnetic and thermal environment of point nuclear spins. In longitudinal, or spin-lattice relaxation

energy from the spin system is transferred to the lattice. Related Articles: Magnetic resonance, Spin-lattice relaxation LCD (liquid crystal display) FIGURE L.21 A crystal structure can be represented as a lattice and a (Diagnostic Radiology) See Liquid crystal display (LCD) motif. Optical density Density Contrast factor LDR (Low dose rate) 547 L ead glass LDR (Low dose rate) occupational exposure is expected to be high, such as during (Radiotherapy, Brachytherapy) See Low dose rate (LDR) interventional radiology procedures, the lead apron should be of L Lead the wrap-around type. A combination of vest and skirt is also pos- (General) sible. In cases when the operator is always facing the radiation source, the apron may be open at the back or with less lead at the back, in order to reduce the weight and cost. Symbol Pb Further Reading: IAEA (International Atomic Energy Element category Group IV metal Agency). 1996. International Basic Safety Standards for Protection Mass number Z various against Radiation and for the Safety of Radiation Sources. Safety Atomic number A 82 Series No. 115, International Atomic Energy Agency, Vienna, Electronic configuration [Xe] 4f14 5d10 6s2 6p2 Austria. Melting point 600.6 K Boiling point 2022 K Lead content (Radiation Protection) Materials used for shielding are often Density near room temperature 11.34 g/cm3 quoted in lead equivalent (or content) which is the thickness of lead which will provide the same amount of shielding. Lead is one of the most used materials for radiation shielding of x-rays Lead (chemical symbol: Pb from the Latin ‘plumbum’) is a and γ rays. The advantage of lead is that it is easily shaped into metal in Group IV of the periodic table. At room temperature and sheets and interlocking bricks. Therefore it is easy to insert lead pressure, lead is a dense solid and possesses a cubic close-packed into wooden panels for the walls (adding protection) or mobile structure (cpp). It is a soft, malleable material, which lends it to shielding barriers. Lead is also used for staff personal protective use in construction. Lead is a poor conductor of heat and electric- devices (lead curtains, aprons, gloves, etc.), patient protection ity compared with other metals. Lead is highly resistant to corro- devices (gonad shield, etc.) or for the shaping of radiation beams sion and is therefore often used to contain corrosive materials such (diagnostic and therapy applications). as strong acids. A flame test identifies lead with a whitish-blue The density of lead is approximately 11 times greater than colour. water. As an indication 4 cm of lead attenuate a Co-60 beam to Isotopes: The four stable isotopes of lead are 204Pb, 206Pb, 207 1/10th of the unshielded value (without considering geometric and Pb and 208Pb. In addition, there are numerous unstable isotopes. beam effects). When other materials are used for shielding, their Medical Applications: The linear attenuation coefficients for respective protective values are usually given in terms of lead lead are high. For example, for 100 keV γ-rays the linear attenu- equivalent. The shielding effectiveness (lead content) of a radia- ation coefficient is 59.7 cm−1. In comparison, the coefficient for copper at this energy is 3.8 cm−1 tion barrier is given in terms of lead equivalent thickness in mm. . Due to the attenuating properties There are also some negative aspects related to the use of lead, of lead it is used as a shielding material in hospital departments as for example that the sheets might slide, due to the high weight where radiation is used, be it in walls around x-ray facilities or (there is need for regular checking) and that the cost is rather high. shielding around vials and syringes in nuclear medicine depart- Therefore, when possible, as in the case of building walls, etc. ments. Lead glass, which contains lead oxide, is used as a concrete is the most recommended shielding material (there are shielding material in situations where the material needs to be also special kinds of concrete). transparent, for example x-ray control rooms or eyewear. In addi- Further Reading: IAEA (International Atomic Energy tion, lead aprons are worn to protect staff working in x-ray rooms, Agency). 1996. International Basic Safety Standards for Protection particularly where fluoroscopic procedures are performed. Gonad against Radiation and for the Safety of Radiation Sources. Safety shields are also made from lead. Furthermore, lead forms part Series No. 115, International Atomic Energy Agency, Vienna, of the material lead zirconate titanate (PZT), a material which Austria. displays a marked piezoelectric effect. As such, PZT is used in the construction of ultrasound transducers. Related Articles: Linear attenuation coefficient, Lead glass Lead drapes (Radiation Protection) Lead drapes or curtains, usually in the Lead apron shape of strips of lead covered with plastic material are used to (Radiation Protection) Registrants and licensees shall ensure that protect the worker who is standing near the patient and the x-ray workers are provided with suitable and adequate personal protec- tube. For example, a typical setting for fluoroscopy investigation tive equipment which meets any relevant regulations or standards. (with the x-ray tube under the patient and the workers nearby) Protective equipment includes lead aprons, thyroid protectors, requires this type of protection to be put between the tube and protective eyewear and gloves. In practice, the need for these pro- the workers. The strips allow movements and adjustments. It is tective devices should be established by the qualified expert on important that the strips are a little overlapping each other, in radiation protection or the radiation protection officer. order to ensure good protection. Regular control should be made Employers, registrants and licensees shall ensure that all per- as lead might slide inside and reduce the protection. sonal protective equipment is maintained in proper conditions and tested at regular intervals. Lead equivalent Gowns, aprons and thyroid protectors are usually made of (Radiation Protection) See Lead content material, such as vinyl, which contains lead. Aprons should be equivalent to at least 0.25 mm Pb, if the x-ray equipment operates Lead glass up to 100 kV and at least 0.35 mm Pb if it operates above 100 kV. (Radiation Protection) Lead glass is a form of glass that has an The aprons should be kept properly, hanging and not folded, and amount of lead in the form of lead oxide. Depending on the amount checked regularly, under fluoroscopy. of lead oxide the atomic number of the glass increases, but at the Several models of lead aprons are available in order to bet- same time maintains the transparency of the glass. Lead glass ter fit the size of the worker and the activity. In situations where usually has an attenuation factor of 1.8–3.2 mm lead equivalent. Lead glasses (eyewear) 548 Leak test Lead glass is used to provide a view port (window) on a shield Lead is also incorporated into rubber or plastic gonad shields. L for ionising radiation, for example a window on a movable lead These articles normally have up to 0.5 mm lead thickness. shield, eyewear, etc. Related Article: Shielding Lead glasses (eyewear) Leadership in medical physics (Radiation Protection) The human eye is particularly sensitive to (General) The issue of leadership is important for all the medical damage caused by ionising radiation (potentially causing cata- and healthcare professions; however, it is of particular importance racts and other opacities). for minor professions such as medical physics. In view of this, There are some work activities with ionising radiation that the European Federation of Organisations for Medical Physics involve workers standing in close proximity to a radiation source, for (EFOMP) organises a leadership module under the umbrella example during interventional radiology/fluoroscopy investigations. of the EUTEMPE project whilst the American Association of In such cases, where it is determined (by a risk assessment) that there Physicists in Medicine has set up a Medical Physics Leadership is a risk of a significant radiation dose to the eyes, the employer is Academy. required to provide protection in the form of leaded eyewear. Related Articles: European Training and Education for Ideally the eyewear should be shaped such that the sides are Medical Physics Experts project (EUTEMPE) protected as much as the front face. However the main problem Further Readings: Caruana, C. J., E. Vano and H. Bosmans. with protective eyewear is that to provide reasonable protection 2015. EUTEMPE-RX module MPE01: Leadership in Medical it may be bulky or heavy such that wearing the eyewear for long Physics, developments in the profession and challenges for the periods feels uncomfortable. Medical Physics Expert (D&IR) in Europe – a first in interna- tional medical physics education and training. Med. Phys. Int. J. Lead gloves 3(2):69; Caruana, C. J., J. A. M. Cunha and C. Orton. 2017. Point- (Radiation Protection) Registrants and licensees shall ensure that counterpoint debate: ‘Subjects such as strategic planning, extra- workers are provided with suitable and adequate personal protec- disciplinary communication, and management have become tive equipment which meets any relevant regulations or standards. crucial to medical physics clinical practice and should become Protective equipment includes lead aprons, thyroid protectors, an integral part of the medical physics curriculum’. Med. Phys. protective eyewear and gloves. In practice, the need for these pro- 44(8):3885–3887. tective devices should be established by the qualified expert or the Hyperlinks: Details of the EFOMP-EUTEMPE leader- radiation protection officer. ship module can be found here: https://eutempe -net .eu /mpe01/; Employers, registrants and licensees shall ensure that all per- Medical Physics Leadership Academy: www .a apm .o rg /or g / sonal protective equipment is maintained in proper conditions str uctur e/ ?co mmitt ee _co de =MP LAWG, https://twitter .com / and be tested at regular intervals. aapmmpla Lead gloves are used to protect the hands, but the weight might limit the movements. In particular, gauntlets are very heavy gloves and should be used only when appropriate (e.g. holding Leak test patients). Special light gloves are available for fluoroscopy. (Radiation Protection) A leak test is an assessment of the integ- Further Readings: IAEA (International Atomic Energy rity of encapsulation of a sealed radioactive source, or from the Agency). 1996. International Basic Safety Standards for Protection housing of an x-ray tube. Leak testing of sealed sources and against Radiation and for the Safety of Radiation Sources. Safety x-ray tubes should each be considered individually, because they Series No. 115, International Atomic Energy Agency, Vienna, involve separate approaches. Austria; Sutton, D. G. and J. R. William, eds. 2012. Radiation Leak testing of sealed radioactive sources is carried out on an Shielding for Diagnostic Radiology. annual basis by a wipe test of the source in question. The aim is to determine whether there is any leakage of the radioactive sub- Lead protection stance from the casing or encapsulation, which would be detected (Radiation Protection) Lead is commonly used to shield against as contamination on the swab. There are statutory limits on the ionising radiation, particularly in the energy ranges where the maximum activity detected using the wipe test method. photoelectric effect is predominant (e.g. x-ray energies below 100 Leak testing of an x-ray tube is performed by measuring dose kV). Attenuation of photons through the photoelectric effect is rates around a completely closed (i.e. collimated) tube-housing, proportional to Z3; hence high Z materials such as lead cause high using sensitive ionisation chambers. Readings are taken at as many attenuation. orthogonal directions around the housing as possible (at least one Lead is used as wall shielding for x-ray rooms. Usually a sheet reading per direction, requiring a minimum of six). X-ray cassettes of lead between 1.3 mm (Code 3) and 2.65 mm (Code 6) sand- may also be employed to further investigate the location of leak- wiched between plywood (to provide rigidity) is used. age from the tube housing. The maximum possible dose rates to Lead is also incorporated into other materials such as glass or bystanders can be calculated from the measured leakage, and com- rubber to provide shielding for other purposes. Lead glass pan- pared to statutory limits. This is particularly relevant with mobile els are used extensively in x-ray rooms to allow the operator to and dental x-ray equipment where operators and/or public may be view the patient during procedures. Radiologists or cardiologists nearby. X-ray tube leak tests are performed as part of commission- making extensive use of fluoroscopy may wear lead impregnated ing tests (including installation

of a replacement x-ray tube). glasses, providing up to 0.5 mm lead protection for the eyes. Related Articles: Wipe test, Leak test (Radiotherapy), Lead rubber is extensively used in the manufacture of x-ray Brachytherapy protective clothing, for example aprons, gloves, thyroid collars. These will normally provide between 0.25 and 0.5 mm lead pro- Leak test tection. This is sufficient to protect against scattered (low energy) (Radiotherapy) Sealed sources used in brachytherapy must be x-radiation. tested for leakage and contamination. In general, yearly checks Leakage current 549 Lens are required. Note that national regulations may vary in their The leakage radiation from a linac is measured in air with the requirements. collimator jaws closed to block the primary beam. A chamber with L For HDR and PDR sources (remote afterloading devices), an appropriate build-up cap is placed at a distance of 1 m from the which are replaced four times a year, the leakage test performed target and also at 1 m from the path of the electrons through the by the manufacturer for each source is stated in the accompanying linac. Measurements are also made in the plane of the patient at source certificates (requirement: leakage and contamination test isocentre height (for details see IEC 60601-2-1). It is possible to <0.185 kBq). The recommended contamination test for the hos- compare this to a measurement made of the primary beam sized pital is an applicator test, performed by placing an applicator in 10 × 10 cm2 at the isocentre for 100 MU. A sufficient reading for a well-type counter to detect any photon emitting contamination. the leakage measurements can be achieved by setting 1000 MU Contamination of the dummy source can be tested by a wipe test, and then dividing the answer by 10 for comparison with the open as it is normally possible to drive the dummy source out using field reading. The average leakage measurements for the region manual control. Note that the dummy source is also replaced at shielded by the collimators should be less than 0.5% of the open regular intervals. field and the average leakage for the region surrounding the accel- For LDR and MDR sources (afterloading and manual load- erator and waveguide should be less than 0.1% of the open field. ing), the recommendation is to check contamination of sources A simple test to confirm the coverage provided by the lead or leakage of radioactive material using a wipe test. Instruments within the head of the linac, in addition to local ‘hot spots’, can used to cut 192Ir-wires should also be checked. be performed by wrapping the head of the linac with radiographic For permanent seed implants, the source certificate states film. The jaws are fully closed and 1000 MU delivered. To allow (Oncura 125-I RAPID Strands): ‘All seeds have passed a leakage comparison some known control films should be generated by and contamination test showing less than 0.185 kBq, 0.005 μCi of irradiating film with 10 and 100 MUs placed at the isocentre with removable Iodine-125 activity’. 1 cm build-up and a field size of 10 × 10 cm2. A comparison of Further reading can be found in the ESTRO Booklet No. 8. the optical density readings can then be performed to assess the Abbreviations: ESTRO = European Society for Therapeutic integrity of the shielding. Radiology and Oncology, HDR = High dose rate and PDR = For high energy linacs (energy greater than 10 MV) it is also Pulsed dose rate. necessary to assess the level of neutron leakage when designing Related Article: Dummy source treatment rooms. Further Reading: Venselaar, J. and Pérez-Calatayud, J., eds. Related Articles: Boron neutron capture, Collimation, 2004. A practical guide to quality control of brachytherapy equip- Treatment head, Maze, Secondary barrier, Tongue and groove ment, ESTRO Booklet No. 8, Brussels, Belgium. leakage Further Reading: IEC (International Electrotechnical Leakage current Commission). 2002. Medical Electrical Equipment – Part (General) Leakage current is an unwanted flow (leak) of current 2-1: Particular Requirements for Basic Safety and Essential along a path different from that intended. In electric circuits and Performance of Electron Accelerators in the Range 1 MeV to 50 components, it is mainly due to insufficient or faulty insulation MeV, IEC 60601-2-1, 2nd edn., Geneva, Switzerland. (e.g. leakage current of a capacitor). In electronic circuits, it is an inherent characteristic of semiconductor devices (e.g. leakage Lens current in analogue switches). (Ultrasound) Mechanical focusing of an ultrasound beam is pos- sible with an acoustic lens or with a curved transducer element, Leakage radiation Figure L.22. An acoustic lens works in a similar way to an opti- (Radiation Protection) Leakage radiation results from the lack of cal lens. However, in the optical case speed of light in the lens integrity of radiation shielding. This may occur from cracks in the material is always lower than speed of light in air. In the acousti- housing of an x-ray tube. Leakage radiation may also refer to radi- cal case, speed of sound in the lens material can be either higher ation penetrating through service-ducts in walls, gaps between or lower than that for tissue (1540 m/s). Convex (speed of sound shielded doors in x-ray facilities, etc. This results in higher than lower than tissue) or concave l (speed of sound higher than tissue) expected dose rates adjacent to the gap in the shielding, with lenses can be used for focusing, Figure L.22. potential consequences for harm of persons in that area. The focal distance, F, is determined by the radius of curvature Leakage radiation from facilities is checked at construction as (=F) in the case of a curved transducer element case and by both part of general tests to assess actual wall-shielding against a build- the curvature and speed of sound in the case of focusing with a specification. Leakage radiation from x-ray units is assessed at lens. Focusing can only be achieved within the unfocused trans- unit commissioning and installation of new x-ray tubes, by means ducer’s near field, with stronger focusing effect in the first half of of a leak test. the near field. The beamwidth WF at a distance F depends on the Related Article: Leak test transducer diameter d and the frequency. It can be approximated to WF = Fλ/d. Leakage radiation The focal zone is defined as the region where the beam width (Radiotherapy) Leakage radiation is the ionising radiation which W < 2WF. has passed through the protective shielding of a radiation source. Mechanical focusing is used in single element transducers, It is important to reduce this leakage to protect the patient from mechanical sector scanners and in the elevation plane for conven- unwanted radiation and to help reduce the necessary amount of tional array transducers. In array transducers beam focusing in additional treatment room shielding. It is minimised by the use of the image plane is achieved by electronic focusing. significant amounts of protective shielding (lead, tungsten, etc.) in Related Articles: Snell’s law, Elevation, Lens coupling, the head of the linac. Beam-former Lens (Eye) 550 Lethal dose Transducer elements–focussing 400 nm and 1.4 µm. Health Phys. 79(4):431–440; ICNIRP. 2004. L Guidelines on limits of exposure to ultraviolet radiation of wave- lengths between 180 nm and 400 nm (incoherent optical radia- tion). Health Phys. 87(2):171–186; ICNIRP. 2013. Guidelines on limits of exposure to incoherent visible and infrared radiation. Unfocussed Health Phys. 105(1): 74–91; ICNIRP. 2013. Guidelines on limits of exposure to laser radiation of wavelengths between 180 nm and 1000 µm. Health Phys. 105(3):271–295; ICNIRP. 2016. A closer look at the thresholds of thermal damage: Workshop report by an ICNIRP task group. Health Phys. 111(3):300–306; Sihota, Focussed Lens Curved transducer element Ramanjit and Radhika Tandon. 2011. Parsons' Diseases of the Eye. Elsevier, India; Snell, R. S. and M. A. Lemp. 2013. Clinical Anatomy of the Eye. John Wiley & Sons. Focussed Zone of best lateral resolution LET (Radiation Protection) See Linear energy transfer (LET) FIGURE L.22 Focusing with acoustic lens or curved transducer ele- Lethal dose ment. (Courtesy of EMIT project, www .emerald2 .eu) (Radiation Protection) In general, the term lethal dose refers to the radiation dose that will cause the death of the individual that receives it. The 50% lethal dose (LD50), defined as the dose that Lens (Eye) causes a mortality rate of 50% in an experimental group within (Non-Ionising Radiation) The lens is a biconvex structure of the a specified period of time, has been adopted as an end-point for eye that focuses light into the retina, and it is made of up to three scoring radiation death. parts: The effect of a dose of radiation will depend on its magnitude, the proportion and/or the part of the body exposed. Single doses • A capsule that incorporates the whole of the lens up to 10 Gy and multiple doses of 2–3 Gy are routinely delivered • Lens epithelium to a limited area of the body in palliative and radical radiotherapy • Lens fibre cells treatment. However, a single dose of only 2.5–5 Gy to the whole body may be sufficient to cause death. At such doses, death is In an adult a lens measures about 10 mm in diameter and 4 mm caused by radiation damage to the haematopoietic system. in thickness. There is a time delay between the radiation damage and the onset of symptoms since it is the mitotically active precursor cells (precursor cells are stem cells that have developed to the stage where they are committed to forming a particular kind of new blood cell) that are sterilised by the radiation. Hence it is the subsequent supply of mature red blood cells, white blood cells and platelets that is reduced, and so it is only when the cur- rent circulating mature cells begin to die off that the inability of the depleted precursor cell population to replace them becomes apparent. In humans, the peak incidence of death from haematological death occurs at about 30 days after exposure but deaths continue for up to 60 days. Therefore LD50 estimates for humans are usually expressed as the LD50/60. Studies of patients who have received total body irradiation for radiation therapy and those involved in radiation accident/incidents such as the victims of Hiroshima and Chernobyl have attempted to estimate the value of LD50/60 and this is generally quoted at 4 Gy. However, there are many factors that influence the response of an individual within a population. For example, females appear generally to have a greater dose toler- ance than males as do young adults compared with the very young and elderly. Abbreviations: LD50 = 50% lethal dose and LD50/60 = 50% lethal dose measured at 60 days from time of exposure. Related Articles: AORD, Eye, Lens, UV light hazard Related Articles: Cell cycle, Fractionation, Palliative treat- Further Readings: Coleman, A., F. Fedele, M. Khazova, P. ment, Radiosensitivity, Tolerance, Total body irradiation Freeman and R. Sarkany. 2010. A survey of the optical hazards Further Readings: Hall, E. J. and A. J. Giaccia. 2006. associated with hospital light sources with reference to the Control Radiobiology for the Radiologist, 6th edn., Lippincott Williams & of Artificial Optical Radiation at Work Regulations 2010. J. Wilkins, Philadelphia, PA; Nias, A. H. W. 1998. An Introduction Radiol. Prot. 30(3):469; ICNIRP. 2000. Revision of the guidelines to Radiobiology, 2nd edn., John Wiley & Sons Ltd., Chichester, on limits of exposure to laser radiation of wavelengths between UK. Life cycle of equipment 551 Lifetime attributable risk (LAR) L FIGURE L.23 Lifecycle of a medical device. Life cycle of equipment Lifetime attributable risk (LAR) (General) The life cycle of equipment describes the succession (Radiation Protection) There is a link between being exposed of the stages that a product passes through, from the moment it is to radiation and developing stochastic effects such as cancer or first conceived to when it is disposed of. hereditary disease. These effects have a latent period and thus As Figure L.23 shows, the equipment life cycle is actually there is a future risk associated with radiation. a series of nested cycles, starting with an idea and passing In the absence of sufficient radiobiological knowledge, quan- through different stages before the placement on the market, titative risks of cancer following exposure to radiation must be namely: obtained from epidemiological studies of suitably exposed groups of humans. • Research and Development

(R&D) cycle usually com- The Biological Effects of Ionizing Radiation (BEIR) VII Phase prises of the design, development, testing and evalua- 2 report produced by the National Academy of Sciences (National tion stages, which lead to a prototype. Research Council) comprehensively reviewed biological and epi- • Preclinical and clinical testing are required to test the demiological data related to health risks from exposure to ionis- prototype, with sub-phases including the planning, the ing radiation (Smith-Bindman et al., 2009). This report provides a testing, the data analysis and the submission for regula- risk model called the Lifetime Attributable Risk (LAR) of cancer tory approval. incidence and mortality. It provides a method to estimate the LAR • Manufacturing and placement on the market, if and of cancer based on the magnitude of a single radiation exposure only if the regulatory approval is obtained and the con- and a patient’s age at the time of that exposure. formity with the legal framework is certified. The LAR is defined as the additional cancer risk above the • Use and procurement, which involve health technol- baseline cancer risk. This can be calculated for specific cancers ogy assessment, reimbursement procedures and also as well as for all cancers combined (Smith-Bindman et al., 2009). equipment management and vigilance, which are usu- It is an approximation of the risk of exposure-induced death ally under the competences of the clinical engineering (REID) model and describes excess deaths (or disease cases) over department. a follow-up period with population background rates determined • Obsolescence, when the equipment can no longer be by the experience of unexposed individuals (ICRP, 2007). repaired, is obsolete and, therefore, dismissed. The LAR for a person exposed to dose D at age e is calculated as follows: Related Articles: Specification of medical device; Standards; Equipment management; Maintenance, Clinical engineering. LAR (D,e) = M(d,e,a)* S(a) /S(e) Further Reading: Pallikarakis, N. 2019. Medical devices regulations, management and assessment; new trends new needs. The summation is from a = e + L to 100, where a denotes attained In International Conference on Nanotechnologies and Biomedical age (years) and L is a risk free latent period (L = 5 for solid can- Engineering, Springer, Cham. cers; L = 2 for leukaemia) (National Research Council, 2006). Light 552 Light localiser S(a) is the probability for surviving until the age of a and S(a)/S(e) radiometer using light-emitting diodes as detectors. Appl. Opt. L is the probability of surviving to age a conditional on survival to 47(36):6753–6762; Ihrke, I., J. Restrepo and L. Mignard-Debise. age e. All calculations are sex-specific; thus, the dependence of all 2016. Principles of light field imaging: Briefly revisiting 25 years quantities on sex is suppressed (National Research Council, 2006). of research. IEEE Signal Process. Mag. 33(5):59–69; Kitsinelis, M(d, e, a) is the EAR which is the additional risk above the S. and S. Kitsinelis. 2015. Light Sources: Basics of Lighting background absolute risk (Wakeford, 2011). Technologies and Applications, CRC Press. Because of the various sources of uncertainty, it is important to regard specific estimates of LAR with a healthy scepticism, Light field placing more faith in a range of possible values (BEIR VII, Page (Radiotherapy) It is essential to be able to confirm the settings 278) (National Research Council, 2006). to be used for creating the treatment field shape and also to visu- Related Articles: Absolute risk, Excess risk alise the treatment field on the patient’s skin. To achieve this, a Further Readings: ICRP. 2007. ICRP Publication 103, The light field which exactly mimics the radiation field to be deliv- 2007 Recommendations of the International Commission on ered can be used. The linear accelerator contains a light bulb Radiological Protection, s.l.: Elsevier Ltd.; National Research within the head and a mirror to shine the field through the jaws. Council. 2006. Health Risks from Exposure to Low Levels of The bulb is aligned such that the light field will mimic the radia- Ionizing Radiation (BEIR) VII Phase 2, The National Academies tion field to be delivered to within 1 mm on each jaw. It is impor- Press, Washington, DC; Smith-Bindman, R., M., M. Jafi Lipson tant that the distance from the bulb to the mirror is the same as and B. Ralph Marcus. 2009. Radiation dose associated with com- the distance from the source to the mirror. The light field also mon computed tomography examinations and the associated life- allows a projection of the crosshairs to be displayed and used to time attributable risk of cancer. JAMMA, J. Am. Med. Assoc. pp. confirm the crosshair stability with collimator rotation. It should 2078–2086; Wakeford, P. R. 2011. Richard Wakeford Modulating be possible to achieve crosshair rotational walkout to be less Factors and Risk Assessment. [Online] Available at: www .icrp than 0.5 mm. .org /docs /Richard %20Wakefor d %20M odula ting% 20Fac tors% It is important that the coincidence of the light and radia- 20and %20Ri sk %20 Asses sment .pdf [Accessed 16 July 2019]. tion fields is checked on a regular basis over a range of different Hyperlinks: Richard Wakeford Modulating Factors and Risk field sizes. The most common method to do this is using a radio- Assessment: www .i crp .o rg /do cs /Ri chard %20Wa kefor d %20M graphic film; however with the increase in digital techniques odula ting% 20Fac tors% 20and %20Ri sk %20 Asses sment .pdf other methods are becoming popular such as using electronic portal imaging devices. A piece of radiographic film is marked Light by pen according to the field indicated by the light field, then (Non-Ionising Radiation) It is the most common term used in without moving the film or changing the field size the film is light therapy or safety to indicate the light received by the stricken irradiated. It is then possible to compare the field size accord- subject/object. This is the light flux through a unit area, and it is ing to both the radiation (readout using a densitometer) and the measured in Wm−2. light field (measured with a ruler), and the coincidence between Skin exposure to solar or artificial light is usually expressed the two. as an irradiance. The calibration of the secondary collimators (jaws) can also be checked by using the light field projection onto a piece of graph paper positioned at the isocentre plane. Related Articles: Collimator, Crosshairs, Optical distance indicator Light guide (Nuclear Medicine) Light guide refers to the connection between the crystal and the photomultiplier tube (PM tube) in a gamma camera. The light photons are generated in the crystal as a result of electrons released in a photon–electron interaction, that is pho- toelectric effect or Compton interaction. The light guide allows photons to pass from the crystal to the PM tubes where the signal is strengthened. In a gamma camera, several PM tubes are packed close together and connected to a single crystal by one light guide. In early emission imaging the light guide was kept relatively wide in order to attain a uniform irradiation of the PM tubes. A wide light guide will have a degenerative effect on the spatial resolution (as illustrated in Figure L.24) because of an increase in deposition point to PM tube distance. However a modern uniformity cor- rection allows for narrower light guides without an appreciable decrease in uniformity. Related Article: Photomultiplier (PM) tube Related Articles: AORD, Light source, Photodiode, Light localiser Phototherapy, Radiance, UV dosimetry (Diagnostic Radiology) A device used to adjust the radiation Further Readings: Czapla-Myers, J. S., K. J. Thome and S. field/beam at an exact place. Radiographic devices use a dia- F. Biggar. Design, calibration, and characterization of a field phragm (light beam diaphragm) to mimic the x-ray field with light Light radiometer 553 Limitations to the MIRD formalism PM tubes PM tubes L Light guide Crystal FIGURE L.24 A wide light guide will have a negative impact on the spatial resolutions because of the increase in deposition point to PM tube distance. and adjust the exposure field over specific parts of the patient. Limitation Scanning devices (as CT scanners) use laser beam mimicking the (Radiation Protection) The third principle of protection against x-ray scanning beam to set up the exact scanning plane. Precise ionising radiation for workers and members of the public speci- positioning of the light localiser is checked during quality control fied by the International Commission on Radiological Protection procedures (Figure L.25). is limitation. Related Article: Diaphragm collimator Once the use of ionising radiation has been justified and optimised the exposure must be limited. The current (ICRP Light radiometer Publication 103) recommendations for exposure limitation are (Non-Ionising Radiation) See Radiometer divided into stochastic limits and deterministic limits. The stochastic limit is 100 mSv effective dose over five years Light yield in scintillation detectors with no more than 50 mSv effective dose in any one year. (Nuclear Medicine) Light yield is an important parameter of a The deterministic limits are organ specific. The annual equiv- scintillating material. It is the measure of the number of light pho- alent dose limits are as follows: tons that are emitted per unit radiation energy deposited in the material. It takes for a number of materials an average of three Eyes 150 mSv times the bandgap to create an electron–hole pair. As an exam- Skin 500 mSv (averaged over 1 cm2) ple, for sodium iodine with thallium impurities 20 eV is required Hands, feet, etc. 500 mSv to create an electron hole pair. If 1 MeV were deposited in the sodium iodine detector, 5 × 104 electron hole pairs would be cre- ated. The total number of light photons created from 1 MeV is 4 Related Articles: Justification, Optimisation, Dose limits × 104 with 3 eV each. The yield is thus close to 1 light photon per electron–hole pair. Limitations to the MIRD formalism Related Articles: Inorganic scintillators, Scintillators, NaI(Tl) (Nuclear Medicine) The MIRD formalism is used when calcu- detector crystal, Bismuth germanate (BGO) lating the total absorbed dose to patients in nuclear medicine Further Reading: Glenn, F. K. 2000. Radiation Detection examinations. The formalism is an important tool to compare and Measurement, 3rd edn., John Wiley & Sons Ltd., Chichester, the absorbed dose between different organs and also between UK, pp. 231–234. patients. It is also a useful tool in the approval process of new radiopharmaceuticals. But there are a number of important limi- tations that users should be aware of. Although fundamentally correct, the absorbed fractions, ϕ, are often based on standard phantoms of the human anatomy and are not specific for each individual in regards to shape, size and orientation of organs. Another limitation to the formalism is the assumption that the activity is uniformly distributed in the organ. Consider a radionuclide that emits low energy particles with low penetra- tion length (e.g. Auger). In such a case the assumption leads to a large underestimation of the local dose in organ volumes with high uptake and consequently an overestimation in volumes with lower uptake. The estimation of the accumulated activity à for new radio- pharmaceuticals can be complicated. The bio-kinetics can be acquired from human data, but such data are not always applicable to individual patients because of the pathophysiological effect on FIGURE L.25 Checking the light localiser of a CT scanner using x-ray uptake, clearance and excretion of the radiopharmaceutical. film in envelope. The MIRD formalism does not involve dose rate. Limited angle tomography 554 L ine pair Related Article: MIRD formalism L Further Readings: Loevinger, R. F., T. Budinger and E. E. Watson. 1988. MIRD Primer for Absorbed Dose Calculations, The Society of Nuclear Medicine, Inc., New York; Snyder, W., M. Ford and G. Warner, G. 1978. Estimates of Specific Absorbed Fraction for Photon Sources Uniformly Distributed in Various Organs of a Heterogeneous Phantom, MIRD Pamphlet No. 5 (revised), Society of Nuclear Medicine, New York. Limited angle tomography (Diagnostic Radiology) Radiographic tomographic procedure (linear classical tomography) in which the angle of motion of the x-ray tube is set to a low value (like 10 angular degrees) to pro- duce relatively thick image slices (as Zonography). Linac (Radiotherapy) See Linear accelerator Line focus principle (Diagnostic Radiology) The line focus principle, first described Linac cone beam CT by Dr. O. Goetze, in 1918, is applied to most x-ray tubes. The (Radiotherapy) Attachment of cone beam (kilovoltage) CT units usual cathode consists

of a helical heated filament mounted in a to commercial radiotherapy linacs has become widespread. Such focusing electrode. The resulting electron beam focused on the units facilitate image-guided radiotherapy by allowing a patient’s anode surface forms a focal spot that is an image of the elongated position and anatomy to be verified directly before their treat- (line shaped) heated filament. The length is generally the largest ment. They also open up possibilities for adaptive radiotherapy dimension of the focal spot and is highly dependent on the angle delivery. of the anode surface and the direction from which the focal spot Related Articles: Cone beam CT, Image-guided radiotherapy, is being observed. For a line focus principle-related diagram see Adaptive radiotherapy the article Stationary anode. Related Article: Stationary anode Further Reading: Goetze, O., DP 370 022. 1918 Line artefact (General) Line artefacts cross the image as stripes or solid or dashed lines. In nuclear medicine they have been caused by Line of response (LOR) problems with the analogue to digital converters. In MRI they (Nuclear Medicine) A registered event is assumed to originate may be caused by RF leakage or errors in the RF transmission. from a decay or annihilation somewhere along a line of response They can also be caused by problems with the image memory (LOR). The way a LOR is determined differs between PET and a array. scintillation camera due to the different spatial localisation pro- Related Article: Artefact cesses involved. For example, in PET imaging, when two opposite detectors simultaneously register an annihilation photon a LOR is ‘drawn’ between them, that is the event is assumed to have Line density occurred somewhere along the LOR. In a scintillation camera, if (Ultrasound) Line density relates to the number of active scan the camera uses a parallel-hole collimator, each hole is associated lines (per unit distance or per unit area) which emanate from with a LOR that is perpendicular to the detector surface. the transducer into the scan (imaging) plane. For linear trans- ducers these can be imagined as a series of parallel vertical Line pair lines propagating perpendicular to the transducer’s scanning (Diagnostic Radiology) A line pair is one line and one adjacent surface. blank space in a resolution or bar-phantom test object as illus- The influence of line density can be used in conjunction trated in Figure L.26. with B-mode, spectral Doppler, or colour Doppler modalities. Where two or more of these applications are used concurrently the values for each can differ with the total number available shared. Resolution test pattern An increase in line density will improve the lateral spatial res- olution of the image but at the expense of a reduced frame rate. A typical frame rate can be up to about 60 Hz. For any given frame cycle the transducer will wait to receive echoes from all active scan lines. As line transmission is staggered across the array the greater the number of scan lines and the longer the duration for each frame. At a fixed line density, the frame rate can be increased by imaging at shallower imaging depths. Similarly, for colour 1 2 3 4 5 6 7 8 Doppler imaging a reduction in the depth and width of the box Spatial frequency (lp/mm) will increase colour Doppler frame rates. However, newer plane wave techniques allow for improved imaging at very higher frame FIGURE L.26 Line pair test pattern (phantom). (Courtesy of Sprawls rates. Foundation, www .sprawls .org) Line scanning 555 Linear accelerator For x-ray imaging applications, the lines consist of an absorb- to the neutral conductor). Line voltage in a three-phase system is ing material such as lead. The typical test pattern consists of a 1.73 times higher than phase voltage. L series of different-sized lines and spaces, or line pairs. Three-phase power systems are used for supplying large power The size of the lines and spaces is specified as the number of scale customers. Some medical imaging equipment, for example line pairs in a unit length, line pairs per mm (lp/mm, also known x-ray machines, CT scanners and MRI systems, are usually pow- as spatial frequency, in analogue to cycles per second – c/s ered by a three-phase electric power system. frequency). Related Articles: Y-voltage, Star voltage Related Articles: Spatial resolution, Detail resolution, Bar phantom Linear accelerator (Radiotherapy) A linear accelerator (linac) is the most common Line scanning megavoltage unit used in radiotherapy to produce MV pho- (Ultrasound) The term line scanning refers to images constructed ton and MeV electron beams at several different energies, see from an ultrasound beam or ultrasound beams. The echoes arise Figure L.27. It is gradually replacing the 60Co and kilovoltage from reflection and scattering along the line of the beam. Single treatment machines due to its increasing range and scope of clini- line amplitude scans are described as A-line scans. B-mode scans cal treatment techniques. It is housed in a gantry that allows the are constructed from echoes along several lines. radiation beam to be rotated isocentrically through 360° about the Related Articles: A-lines, B-lines, B-mode treatment couch. An electron gun is used to inject electrons into a waveguide, Line source model in synchrony with radiofrequency energy from a magnetron or (Nuclear Medicine) This refers to the process in nuclear medicine klystron. This accelerates the electrons to high energies and they whereby a line source is used to examine the spatial resolution of are then magnetically steered within the treatment head to either the imaging system. form an electron beam or directed at a target to produce brems- A planar image of one or more line sources is acquired under strahlung radiation, often incorrectly called an x-ray beam. certain specified conditions (e.g. according to NEMA protocol). The structure and main components of a linac can be seen in A profile is then drawn across the source. The full width at half Figure L.28, and are as follows: maximum (FWHM) of this curve gives the spatial resolution of the system. • Electron gun Related Articles: Spatial resolution, Full width at half maxi- • Accelerating waveguide mum (FWHM) • RF power source • Pulsed modulator Line spread function (LSF) • Water cooling system (Nuclear Medicine) A point spread function (PSF) is a function • Vacuum system that describes an image system degrading effect due to inher- • Steering and focusing magnets ent limitations. In its basic concept, a point spread function is a • Bending magnets 2D (or 3D) image of an infinitesimally small object. In nuclear • Treatment head medicine applications this will be a point source that is much smaller in radius compared to the expected spatial resolution of Electron Gun and Waveguide: Electrons are produced in the system. the electron gun by thermionic emission from a heated tungsten Sometimes it can be difficult to work with and prepare a cathode and focused into a central stream by accelerating anodes. point source. A measurement of the line spread function (LSF) They pass into a waveguide in which they are accelerated up to can therefore be the choice. The LSF is the line spread function speeds close to the speed of light. As the electrons travel down the integrated in either x- or y-direction. In practice a LSF measure- waveguide, their magnetic fields interact with the RF field (elec- ment of a line source is made with a scintillation camera. The tric field component aligned along the long axis) and they experi- line source is usually a capillary plastic tube with a very small ence a force, accelerating through the tube. By using travelling diameter filled with a radioactive solution. The source should be RF waves, or standing RF waves, the force continues to accel- longer than the tails of the expected PSF. The LSF function is erate the electrons even as they move, producing a high-energy then approximated by a profile through the centre of the imaged line source. Abbreviations: LSF = Line spread function and PSF = Point spread function. Related Article: Point spread function, MTF Line voltage (General) The voltage provided by a power line and measured at the point of use. Households are usually connected to a single-phase elec- tric power system. Line voltage is measured between the active (phase) and neutral conductor. The nominal value of the line volt- age and the nominal frequency of the alternating current differ geographically. In Europe and Asia it is in most countries 230 V/50 Hz, and in the Americas 115 V/60 Hz. In a three-phase electric power system, line voltage is mea- sured between any two active (phase) conductors (and not referred FIGURE L.27 Linear accelerator and treatment couch. Linear and shift-invariant systems 556 Linear and shift-invariant systems L Vacuum system Steering coils Focusing coils Steering coils Electron gun Electron beam transport Target Accelerating waveguide Exit window Primary collimators Gas pressure Pulsed system Flattening filter modulator Circulator Dual ion chamber Microwave Water Upper power cooling jaws source system Lower Control unit Multileaf collimator jaws FIGURE L.28 Schematic diagram of an x-ray linac. (From Podgorsak, E. B., Review of Radiation Oncology Physics: A Handbook for Teachers and Students, International Atomic Energy Agency, Vienna, Austria, 2003.) electron beam. An advantage to the standing waveguide is that it Abbreviations: linac = Linear accelerator and RF = produces a higher accelerating gradient per metre, allowing for Radio-frequency. more compact vertically designed machines that need no bending Related Articles: Treatment head, Magnetron, Klystron, magnets. Steering and focusing magnetic coils are used along the Bending magnet, Waveguide waveguide to counter the dispersal of the electrons from the cen- Further Reading: Podgorsak, E. B. 2003. Review of Radiation tral axis. This is all contained within a vacuum system to ensure Oncology Physics: A Handbook for Teachers and Students, that the accelerated electrons are not deflected from their path by International Atomic Energy Agency, Vienna, Austria. collisions with gas molecules. RF Power Source: There are two main RF power sources used Linear and shift-invariant systems in linacs – magnetrons and klystrons. A magnetron is an RF oscil- (Diagnostic Radiology) A system is linear if superposition holds. lator that extracts energy from electrons in a resonant structure Superposition refers to the ability of a system to process signals within a magnetic field. A klystron is a tuned RF amplifier, which individually and then sum them up to process all the signals amplifies an external source of RF to higher power. Klystrons are simultaneously. Formally, suppose that two bi-dimensional sig- generally considered to be more reliable than magnetrons, which nals (e.g. images) f and g are present at the input of system T{ }. can be affected by the earth’s magnetic field. However they are When applied to the system individually they produce: much larger and need more sophisticated waveguides to trans- fer the power, as they cannot be mounted within the gantry, like f1 (x, y) = T { f ( x, y)} magnetrons. Treatment Head: The target used to create an MV photon beam is made from Tungsten, and is placed in the path of the elec- g1 ( x, y) = T {g ( x, y)} tron beam. The Bremsstrahlung photons produced by the interac- tions in the target will be highly peaked along the central axis. To If superposition holds, for arbitrary constants a and b, it is true create a clinically useful beam, a flattening conical filter is used that: to reduce the intensity in the centre. This filter is energy specific. A dual ionisation chamber system is used to monitor energy, flat- ness and dose. T {af ( x, y) + bg ( x, y)} = aT { f ( x, y)} + bT {g ( x, y)} If an electron beam is required, the photon target is moved out of the path of the electron beam, and is replaced by two scattering = af1 ( x, y) + bg1 (x, y) foils. These modify the shape and spectrum of the intense pencil beam of electrons to produce a clinically useful beam. Electron If superposition does not hold, the system is nonlinear. Practically, beams require additional collimation called applicators to reduce this means that distortion or artefacts are introduced by the sys- the dose from the high scatter that occurs in air. The beam current tem for some input signals. Typically, a range of linearity is

com- in electron mode must be reduced by a factor of 100 (along with monly associated with the system. the frequency of the RF source) from that of x-ray mode to prevent A system is shift-invariant if its properties or characteristics dangerously high dose rates for electrons. do not change with the position. Formally, given Linear array 557 Linear Boltzmann transport equation (LBTE) solver g ( x, y) = T { f ( x, y)} L and a shift offset (x0, y0), the shifted input f1 ( x, y) = f ( x - x0, y - y0 ) must produce system output g1 ( x, y) = g ( x - x0, y - y0 ) An example of a shift-varying system is an adaptive image pro- cessing filter where its effect changes within the image according to some local properties. Engineers and scientists are typically most interested in work- ing with systems that are both linear and shift-invariant because such systems can meet demanding real-world requirements and simple mathematical tools exist for their analysis in the time, frequency and other transformed domains. These tools include FIGURE L.30 Linear array image of a common carotid artery. Beam the impulse response function (IRF). The IRF is used to develop steering of the colour flow and spectral Doppler beam helps to achieve a general input/output relationship for the system. Knowing the good beam/flow angles. IRF, the system output for any input is the convolution integral with the impulse response. Related Articles: Impulse response function, Convolution transducers, elements are subdivided across the width of the integral probe. Related Articles: Matrix array, Curvilinear array, Transducer array Linear array (Ultrasound) Linear arrays are array transducers in which the elements, typically 64, 128 or 256 are arranged in a straight line Linear array transducer (Figure L.29). This image format is rectilinear with the width dic- (Ultrasound) See Linear array tated by the length of the array. Linear arrays are most commonly used for imaging of superficial tissue including breast, thyroid Linear attenuation coefficient and testes, musculoskeletal applications and peripheral arter- (Diagnostic Radiology) Attenuation of photonic radiation is an ies and veins. They typically use frequencies in the range 3–16 exponential process that can be represented by the equation: MHz depending on the application. For colour flow and pulsed wave spectral Doppler imaging, linear arrays enable electronic I = I e-mx o beam steering (Figure L.30) which permits an adequate beam/ flow angle for peripheral vessels, many of which run parallel or where nearly parallel to the skin surface. In some systems, a trapezoid Io is the initial beam intensity image can be obtained using the outer elements for beam steering I is the final beam intensity in B-mode. x is the distance travelled Multi-D or matrix linear arrays are offered by some manu- and μ is the attenuation coefficient of the material (absorber) facturers with dynamic focusing in the elevation plane. In these The coefficient μ normally refers to the linear attenuation, although it could equally refer to the mass or atomic attenuation coefficient. It is measured in m−1 (for practical reasons – mm−1). In x-ray imaging linear attenuation coefficient of various tis- sues determines their visualisation – hence the contrast of the image. The linear attenuation coefficient varies with the energy of the radiation (Figure L.31). This variation determines the opti- mal energy (e.g. kV) for visualisation of specific anatomical struc- tures. See Attenuation. Related Articles: Attenuation, Mass attenuation coefficient, Atomic attenuation coefficient Linear Boltzmann transport equation (LBTE) solver (Radiotherapy) The linear Boltzmann transport equation (LBTE) may be solved deterministically (rather than stochastically as in FIGURE L.29 Linear arrays. A modern high frequency transducer (left) the case of Monte Carlo algorithms), in order to determine the is shown next to an older low frequency (3 MHz) linear array (right) used macroscopic behaviour of ionising radiation passing through a for abdominal and obstetric imaging. medium as relevant to radiotherapy. Linear (classical) tomography 558 Linear (classical) tomography L FIGURE L.31 Dependence of linear attenuation coefficients (for bone, muscle, fat) from kV. The dimensionality of the LBTE when applied to real-world situations is considered to render it impossible to solve analyti- Focal spot cally, and so LBTE solvers rely on numerical methods to reach their solutions. At each spatial point in the patient, there exist pho- ton and electron fluences (discounting any other particle types) at a range of energies, in a range of directions, all of which need to be transported. LBTE solvers utilise discretisation of this phase space and its Pivot point associated integrals, such that solutions are calculated and given Pivot plane in terms of discrete rather than continuous energies and directions. As an example of how this could be practicable, Bremsstrahlung photons generated in the target in a linac head are forward peaked. Imaged area Body section Therefore, the discretised angles could be focused tightly in this forward direction, with sparser spacing employed elsewhere. Related Articles: Boltzmann transport equation Further Reading: Bedford, J. L. 2019. Calculation of absorbed dose in radiotherapy by solution of the linear Boltzmann transport Receptor equations. Phys. Med. Biol. 64(2). FIGURE L.32 The basic components of a tomographic system showing Linear (classical) tomography the relationship of the tube and receptor to imaged area. (Courtesy of (Diagnostic Radiology) Linear tomography is a radiographic Sprawls Foundation, www .sprawls .org) procedure for producing tomograms or images of selected slices within a patient’s body. The process is illustrated in Figure L.32. During the exposure the x-ray tube and the receptor are moved as of cut, without blurring the image in the slice itself. How this is shown. This motion blurs the anatomical structures both above achieved is illustrated in Figure L.33. and below the slice that is being imaged. The image of an object located that is in the plane of cut, or The tube and receptor are on opposite ends of an arm that tomographic slice, does not move relative to the receptor and is pivots about a point that is in the same plane as the slice being therefore not blurred. However, the image of an object that is not imaged. The location of the pivot point relative to the body can in the plane of cut will move relative to the receptor and will be be adjusted to produce images in different planes, or at different blurred. An illustration is seen on Figure L.36. depths, through the body. The amount of blurring is determined by both the distance of The objective is to blur and reduce the contrast of the anatomi- an object from the plane of cut and the angle through which the cal structures above and below the tomographic slice, or plane tube and receptor is moved as illustrated in Figure L.34. Linear dose response curve 559 Linear dose response curve Focal spot a b L B Object 2 Object 1 X Plane of cut FIGURE L.35 Typical linear tomography x-ray set with linear move- ment of the tube along the axis of the patient table. b a Receptor Br FIGURE L.33 The relationship of object-point blurring to the loca- tion of objects within the body. (Courtesy of Sprawls Foundation, www .sprawls .org) 60° 30° Blur Plane of cut Cut thickness FIGURE L.34 The relationship of object point blurring to distance from the plane of cut and the angle. (Courtesy of Sprawls Foundation, www .sprawls .org) FIGURE L.36 Typical linear tomography of the lung, presenting a bet- ter image of a structure in the lungs with decreased superimposition of the ribs. Linear tomography does not produce a slice with a precise thickness in which there is no blurring. As illustrated here, the blurring increases with distance from the plane of cut. The slice introduction of computed tomography and other scanning imag- thickness in which there is relatively little blurring is determined ing methods the importance of the classical linear tomography by the angle. The angle through which the tube and receptor has rapidly decreased (Figures L.35 and L.36). moves is an adjustable factor and is used to set the slice thickness. The need to produce very thin slices requires high angle of Linear dose response curve movement (long pathway of the tube and film, Figure L.35). (Radiation Protection) The linear dose response curve is just one When this is not achievable technologically the x-ray tube and the example of a dose response curve, together with non-threshold film move around the object not in a linear pathway, but form a dose response curve, non-linear response curve, etc., which may complex curve. This can be a circle (the tube and the film rotate be used either separately or in combination as models to describe in opposite directions), or other complex curves as hypocycloids, the response of the human body to exposure to various types on spiral, etc. Due to the fact that in such complex movements the ionising radiation from both internal and external exposure, and length of the pathway of the tube and the film is much longer at high and low doses and dose rates. than in linear movement, the tomographic slices are very thin. The current internationally accepted framework for radiation However in this case the patient dose is much higher. With the protection is based on a model of potential harm from exposure Linear energy transfer (LET) 560 Linear quadratic (LQ) model Further Reading: ICRP. 2008. Recommendations of the L International Commission on Radiological Protection, Ann. ICRP, ICRP Publication 103, 37(2–4). Linear no-threshold model (Radiation Protection) The basis of radiation protection, as Linear, no threshold described by the International Commission for Radiological (LNT) model Protection (ICRP) is that the stochastic effects of ionising radia- tion can occur at any level of exposure. In other words, radiation Dose exposure can cause harm to humans even for the smallest dose received. Furthermore, it is assumed that the level of risk – that is the FIGURE L.37 The linear no-threshold model. likelihood of suffering a stochastic effect – is proportional to the dose received. This hypothesis is known as the linear no-threshold model, to ionising radiation called the linear no-threshold model, which and can be described by the non-threshold dose response curve assumes a non-threshold linear dose response curve. This model in Figure L.38. suggests that at any level of received radiation dose, harm may The graph demonstrates a linear (i.e. proportional) relation- be caused – that is, a cancer may be induced. The risk of harm ship between dose and risk of effect, and the intercept with the (stochastic effects) is assumed to be proportionate to the dose axes is at the origin – there is no threshold dose below which received (i.e. linear with dose with no threshold for the effect). a human is safe from the risk of harm. However, the epidemio- This is described in Figure L.37. logical evidence used to support the LNT model is mainly at high However, although it is hypothesised that the response is linear doses/high dose rates, and it is therefore merely an assumption with dose, and that there is a risk at even the smallest dose (i.e. that there is a linear relationship down to the smaller doses expe- there is no threshold), there is no evidence to suggest that this is rienced in occupational and diagnostic medical exposures. indeed the case. More recent epidemiological evidence would suggest that the Related Articles: Dose response curve, LNT model linear no-threshold model may be too simplistic, and that there may even be a beneficial effect of exposure to ionising radiation Linear energy transfer (LET) at low doses – this is termed hormesis – the hormetic effect of (Radiation Protection) Linear energy transfer describes the way radiation exposure. in which the energy of incident ionising radiation is transferred to Related Articles: Hormesis, International Commission for the medium through interactions (ionisations) with electrons or Radiological Protection (ICRP), Stochastic effects atomic nuclei. Ionising radiation with low density or sparse inter- Further Reading: ICRP. 2008. Recommendations of the actions along the tracks of incident photons/particles is called low International Commission on Radiological Protection, Ann. LET radiation. Conversely, radiation with a high density of inter- ICRP, ICRP Publication 103, 37(2–4). actions along the track of incident photons/particles is called high LET radiation. Linear quadratic (LQ) model Examples of low LET radiation include gamma rays,

x-rays (Radiotherapy) The linear quadratic (LQ) model is a dose– and beta particles. Examples of high LET radiation include alpha response model and is probably the most widely used and the best particles and neutrons. This property of ionising radiation is currently available for describing the form of the cell survival partly described by the radiation weighting factor used to calcu- curve. It is a second-order polynomial function with only two late equivalent dose. adjustable parameters, α and β, and a zero constant term so that Related Articles: Energy deposition, Radiation weighting fac- the surviving fraction is equal to 1 at zero dose. tor, Equivalent dose The LQ model assumes that there are two components of cell killing by radiation: one proportional to dose and one Linear gradient proportional to the square of the dose. It was first applied (Magnetic Resonance) See Gradient linearity as an empirical description of how the survival of cells var- ied with radiation dose but radiobiological mechanisms were Linear no-threshold dose response (Radiation Protection) The current internationally accepted framework for radiation protection is based on a model of poten- tial harm from exposure to ionising radiation called the linear no- threshold model. This model suggests that at any level of received radiation dose, harm may be caused – that is, a cancer may be induced. The risk of harm (stochastic effects) is proportionate to the dose received (i.e. linear with dose with no threshold for the effect). This is described in Figure L.37. However, although it is hypothesised that the response is linear with dose, and that there is no threshold, there is no strong evi- dence to suggest that this is indeed the case. Dose Related Articles: Linear dose response, Hormesis, Dose response model FIGURE L.38 The linear no-threshold model. Probability of effect Probability of effect Linear quadratic (LQ) model 561 Linear quadratic (LQ) model subsequently attached where the linear component is attributed lethal-potentially lethal (LPL) model of Curtis (1986) is more to single-track events and the quadratic component attributed to appropriate to describe the response with large fraction sizes such L two-track events (for more details see the article on Alpha beta as those used in stereotactic radiosurgery. Guerrero and Allen Li ratio). The LQ model expression for the cell survival curve is (2004) have proposed a modified LQ model, an extension of the given by Equation L.3 in which the surviving fraction, SF, is conventional LQ model, to more accurately describe high-dose the fraction of cells surviving a single dose D and α and β are regimes. constants. The LQ model does not adequately describe the cellular response to radiation at low doses, below about 1 Gy. It has been - aD+b 2 , = ( D Surviving fraction SF e ) (L.3) shown that many mammalian cell lines exhibit hypersensitivity below about 10 cGy, characterised by a cell survival curve slope The LQ model is a second-order polynomial function relating the considerably steeper than that expected by extrapolating back the surviving fraction to dose using only two adjustable parameters, response from high-dose measurements, Figure L.40. A review α and β. of the evidence and possible mechanisms for low-dose hyper- The parameters α and β (with units of 1/Gy and 1/Gy2 respec- sensitivity can be found in the papers by Joiner et al. (2001) and tively) determine the ‘bendiness’ of the survival curve, see Marples and Collis (2008). The LQ model has been modified to Figure L.39. The ratio α/β has the unit Gy, and in a semi-log plot take account of this phenomenon resulting in the induced repair of SF(D) it is the dose where both the linear and the quadratic model, Equation L.4. components of the survival curve are equal. As shown in Figure Surviving fraction, L.39, the response of cells to densely ionising high-LET radiation (e.g. neutrons and α-particles) usually results in a steep, almost ìï é æ a ö æ D ö ü ) í - b 2 ï (L.4 exponential survival curve (see also the article on Relative bio- SF = exp -a D + s ù r ê1 ç -1÷ ´ exp a ç - ÷ú D ý logical effectiveness). In the LQ model description this would be îï ë è r ø è DC øû þï explained by a high α/β ratio. The LQ model has been extended to fractionated radiotherapy DC ~ 0.2 Gy with the derived biological effective dose (BED) widely used for At very high doses (D ≫ DC), equation → LQ model with comparing fractionation schedules and for calculating alterna- parameters αr and β tive regimes, for example to correct for an unwanted interrup- At very low doses (D ≪ DC), equation → LQ model with tion of treatment. Extensions to the BED formulation have been parameters αs and β developed to account for factors such as the incomplete repair of sub-lethal damage and repopulation which can occur during frac- The induced repair model is a modification of the LQ model to tionated regimens (for more details, see the article on Biological account for the hyper-radiosensitivity observed at very low doses. effective dose). The LQ model also requires modification to account for the The LQ model generally works well in describing the response effect of dose rate on response. As stated earlier, the standard LQ to radiation both in vitro and in vivo for the doses typically used model generally works well for treatment schedules utilising mul- in treatment schedules utilising multiple daily fractions. However, tiple daily fractions at the dose rates used clinically for external the LQ formulation results in continuously bending cell survival beam radiotherapy. In such cases, dose delivery takes no more curve which does not match experimental observation if survival than a couple of minutes. However, as dose rate is lowered, the curves are determined for very high doses. In such cases, the time taken to deliver the radiation dose is extended and it becomes dose–response relationship approximates to a straight line in a possible for the radiation response to be modified as a result of the log-linear plot, that is cell killing is an exponential function of dose. It has been proposed that a more kinetic model such as the 1 1 αD αr βD2 αs High LET X-rays 1 Gy Dose α/β FIGURE L.40 Many mammalian cells have been shown to exhibit Dose hyper-radiosensitivity characterised by a considerably steeper slope, αs, than that expected by extrapolating back the response from high-dose FIGURE L.39 The LQ model provides a description of the continually measurements, αr. (Adapted from Joiner, M.C. et al., Int. J. Radiat. downward bending form of the cell survival curve. Oncol. Biol. Phys., 49, 379, 2001.) Surviving fraction Surviving fraction Linear-quadratic dose–response curve 562 Liquid chromatograph following processes: repair of sub-lethal damage, redistribution, Linear stopping power L repopulation and reoxygenation. Such considerations are required (Radiation Protection) The energy lost from a beam of charged in LDR brachytherapy and further details on the effect of dose particle (e.g. alpha or beta) ionising radiation per unit distance rate on response can be found in the article on Dose rate depen- travelled through a medium is known as the linear stopping power dence. The most widely used modified LQ model in such situa- of the material traversed. tions is the incomplete repair model of Thames (1985) shown in Related Article: Stopping power Equation L.5: E = aD + bD2g (t ) (L.5) Linearly polarised (LP) (Magnetic Resonance) If the B1 field vector (i.e. the magnetic where component of the transmitted RF pulse) at a point in space points in a constant direction then the field is said to be linearly éëmt -1+ exp(-mt )ù polarised. g (t ) = 2 û ( t )2 A linearly polarised B1 field is physically equivalent to the m summation of two circularly polarised fields propagating in the same direction but with opposite directions of rotation. E is the level of biological effect α For example, the linearly polarised wave shown in Figure L.41 and β are the LQ model parameters can be written as g depends on half-time for recovery (T1/2) and duration of exposure (t) where B = B1coswty 0.693 m = This can be decomposed into T1/2 The incomplete repair model is a modification of the LQ model B = ½B1( coswty + sin wtx) to account for the effect of low dose rate on radiation response. A comprehensive review of the various parameterisations of + ½B1( coswty - sin wtx) the LQ model used in radiotherapy can be found in the book by Dale and Jones (2007). where the two terms represent vectors rotating with opposing Abbreviations: BED = Biological effective dose, LDR = Low sense. dose rate, LET = Linear energy transfer, LPL = Lethal-potentially In a linearly polarised RF excitation pulse only the circularly lethal, LQ = Linear quadratic and SF = Surviving fraction. polarised component with the same sense as the direction of pre- Related Articles: Alpha beta ratio, Brachytherapy, Cell cession is effective in exciting spins. survival curve, Dose rate dependence, Dose response model, Linear energy transfer, Low dose rate (LDR), Radiosensitivity, Liquid chromatograph Relative biological effectiveness (RBE), Redistribution, Repair, (Radiation Protection) Liquid chromatography is a chemical Repopulation, Stereotactic radiosurgery, Surviving fraction analytical technique based on the separation of a compound, Further Readings: Curtis, K. H. 1986. Lethal and poten- for example radiopharmaceutical, in different components (mol- tially lethal lesions induced by radiation: A unified repair ecules) when passing it through a layer (solid or gel) with the use model. Radiat. Res. 106:252–270; Dale, R. and B. Jones. 2007. of some solvent (liquid phase) as an eluent. This technique is simi- Radiobiological Modelling in Radiation Oncology, British lar to filtration but in this case the separation of the mixture is Institute of Radiology, London, UK; Guerrero, M. and X. Allen a consequence of molecular interactions among the components Li. 2004. Extending the linear-quadratic model for large frac- of a compound and the layer (membrane or column) producing tion doses pertinent to stereotactic radiotherapy. Phys. Med. Biol. different i transit times, that is some molecules (components) 49:4825–4835; Hall, E. J. and A. J. Giaccia. 2006. Radiobiology pass through the layer faster than others. Gel chromatogra- for the Radiologist, 6th edn., Lippincott Williams & Wilkins, phy can be used for separating proteins of different molecular Philadelphia, PA; Joiner, M. C., B. Marples, P. Lambin, S. C. weights, drugs, etc. or for detecting and separating impurities in Short and I. Turesson. 2001. Low-dose hypersensitivity: Current radiopharmaceuticals. status and possible mechanisms. Int. J. Radiat. Oncol. Biol. Phys. 49:379–389; Marples, B. and S. J. Collis. 2008. Low-dose hyper-radiosensitivity: Past, present, and future. Int. J. Radiat. y Oncol. Boil. Phys. 70:1310–1318; Steel, G. G. 2002. Basic Clinical Radiobiology, 3rd edn., Arnold Publishers, London, UK; Thames, H. D. 1985. An ‘incomplete-repair’ model for sur- x B1 vival after fractionated and continuous irradiation. Int. J. Radiat. z Biol. 47:319–339. Linear-quadratic dose–response curve (Radiotherapy) The linear-quadratic model is the dose–response model commonly used to describe the shape of cell survival curves. For further information see the article on Linear qua- dratic (LQ) model. Direction of propagation Related Articles: Alpha beta ratio, Cell survival curve, Dose response model, Linear quadratic (LQ) model, Surviving fraction FIGURE L.41 Linearly polarised B1 field. Liquid crystal display (LCD) 563 Liquid crystal display (LCD) There are three general kinds of liquid chromatography: perpendicular to the molecules. Therefore, light, which is polar- ised in the direction of the long axis of the molecules is absorbed L 1. TLC (thin layer chromatography): A layer of 0.01–2.00 whilst light polarised along the short axis is not. mm, for example silica gel is put on a glass or plastic The simplest form of an LCD is constructed from a thin layer plate. The separation results from a slow penetration of of an organic compound whose cylindrical molecules tend to line this layer by the solution under examination. up parallel to each other. The organic compound, or liquid crystal 2. Column chromatography: A layer is introduced in the (LC), is held between two parallel glass substrates which have column and then the solution passes through it. An had a thin, transparent, layer of metal oxide deposited upon them, example is HPLC (high-performance liquid chromatog- which act as electrodes. On either side

of the substrates is placed raphy) that is used for the purification and identification crossed polarisers, polarisers placed at 90° from each other. Thin of chemical compounds, for example radiopharmaceu- scratches on the glass substrate align the molecules at each sur- ticals. In this technique the phase polarity is significant. face in the direction of polarisation, creating a twisted nematic The sample is injected to the column and then eluent (TN) array. The TN twists the plane of polarisation of light so it is pumped under high pressure (higher than 100 atm). is able to pass through both crossed polarisers, see Figure L.42. The concentration of the different components of the To change the degree of twist within the LCD, electrodes are eluate that are separated by the column is measured placed on the glass substrate, and when a potential difference by a detector. The detector response is proportional to is applied across the LC an electric field is created causing the the concentration that is plotted as a function of time. molecules to un-twist and become almost perpendicular to the The detectors used can be a simple UV–VIS spectro- substrates so that light can no longer pass through both polarisers, photometer (measures the absorbance), a fluorescence see Figure L.43. spectrometer, a mass spectrometer, a laser scatter- To create a monochrome image an array of liquid crystal cells ing spectrometer or a radiation detector, for example is created and each LC cell is backlit, meaning a light source NaI(Tl) for measuring radioactivity of every component is placed behind it. Each cell appears bright when no voltage for radioactive samples. is applied across it (Figure L.42); as the electric field increases 3. Electrophoresis: For example, CE (capillary electro- the molecular array moves and proportion of light able to pass phoresis) is like a thin layer chromatography but the through both polarisers decreases until a maximum voltage is separation is made by the application of an electric field. applied and the cell appears black, Figure L.43. By altering the The sample is introduced to the inlet of a capillary and voltage all greyscales of an image can be reproduced. then the high voltage, for example 30 kV is switched on. This technique makes possible separation of anions and cations in organic and non-organic salts. The detector is Vertical polariser Horizontal polariser placed at the outlet of a capillary. Glass substrate Glass substrate Liquid crystal molecules Abbreviations: CE = Capillary electrophoresis, HPLC = High-performance liquid chromatography and TLC = Thin layer chromatography. Related Article: Chemical exchange Further Reading: Serdyuk, I. N., N. R. Zaccai and J. Zaccai. 2007. Methods in Molecular Biophysics Structure, Dynamics, Function, Cambridge University Press, Cambridge, UK, pp. 112, V 127, 151–152, 155, 168, 391, 409. Switch Field off Liquid crystal display (LCD) FIGURE L.42 A simple twisted nematic, liquid crystal cell used in dis- (Diagnostic Radiology) A liquid crystal display or LCD is an play technology. When no field is applied the cell allows light to pass electronic display that uses cells of liquid crystal molecules to through both polarisers and the cell is ‘on’. create pixels. By varying the electric field across the liquid crystal cells the opacity of each cell can be changed. If the cell is backlit the change in opacity causes a change in the viewed brightness of the pixel. Vertical polariser Horizontal polariser Glass substrate Glass substrate Active matrix flat panel thin film transistor (TFT) LCDs are Liquid crystal molecules used widely in modern computing, and high resolution displays are used in diagnostic radiology for viewing clinical images. These displays use individual liquid crystal cells to form pixels, which are controlled by an active matrix. The term liquid crystal is used to describe a substance in a state between liquid and crystal. It is a liquid in which the mol- ecules exhibit long-range order, a periodic pattern of atomic V positions that extends over many atoms. A nematic liquid crystal Switch is a substance that has different physical and optical properties Field on in different directions because of the spatial anisotropy (elon- gated molecules) of the molecules, which are aligned in a regu- FIGURE L.43 A simple twisted nematic, liquid crystal cell used in dis- lar chain. This anisotropy of the molecules leads to birefringent play technology. When voltage is applied across the cell the molecules properties whereby it has different refractive indices parallel and untwist and the cell is ‘off’. Liquid flow counting 564 Liquid scintillation (LS) counting The TN LCDs were the first type of liquid crystal display to radiation of energy less than 20 keV, β− rays of energy 15–2000 L be developed and their use was restricted by low contrast ratio, keV and α-emitters. narrow viewing angle and slow response time. To overcome these The counting efficiency ɛcount defined as issues different manufacturers have developed LCDs based upon ɛcount = (number of pulses recorded)/(number of radiation several different molecular alignment and electrode patterns. quanta incident on detector) These are twisted nematic (TN), in-plane switching (IPS) and is equal ≈100%. vertically aligned (VA) LCD. The radioactive sample is mixed with liquid scintillator, Related Article: Active matrix liquid crystal flat panel display placed in a sample holder (vial) transparent to optical radiation and then placed between two PM tubes (Figure L.45). The pulses Liquid flow counting registered by PM1 and PM2 tubes pass to a coincidence system. (Nuclear Medicine) Liquid flow counting can be used as an on- The coincidence system allows passing to an amplifier only pulses line monitoring of a radioactive solution flowing through a plastic registered simultaneously by PM1 and PM2 (deriving from the tube, for example in experiment set-up with column chromato- scintillation of the same molecule). The PHA analyses its height graphic techniques. Different radiation detectors, such as Geiger– (amplitude) and the result is presented in the recording device in Müller, CdTe, NaI(Tl) may be used depending on the application analogue or digital form. The LS equipment is calibrated with an and the decay characteristics of the radionuclide. The detector can isotope with known energy of particles and in this way it is pos- be connected to a single- or multi-channel analyser. sible to identify measured radioisotopes and its quantity in the Further Reading: Knoll, G. F. 1999. Radiation Detection and sample. Measurement, John Wiley & Sons, New York, pp. 1–802. The detection output strongly depends on counting quench- ing. Quenching can be caused by interference in optical photons Liquid metal bearing energy transfer by the sample and liquid scintillator (chemical (Diagnostic Radiology) One of the latest x-ray tubes with rotating quenching), by dilution of the sample in scintillation solution anode uses special spiral groove bearings using as lubricant liq- (dilution quenching) or by coloured substance (colour quenching). uid metal (eutectic alloy of Gallium, Indium and Tin) which has There are special experimental methods to take the quenching melting point of −10°C. At room temperature this alloy is liquid into account and to obtain accurate counting of measured sample. as mercury, but has lower vapour pressure. This tube is patented Chemoluminescence can also be a problem. by PHILIPS®. The β− radioisotopes often detected by LS counting include: The spiral groove bearings (Figure L.44) have special profile H-3, C-14, P-32, S-35. (as in car tyres), which allows very smooth rotation over the ‘liq- Abbreviations: LS = Liquid scintillation, PHA = Pulse height uid metal’, using an effect similar to the familiar ‘aqua-planning analyser and PM = Photomultiplier. effect’. This allows for a very low friction (i.e. excellent rotation) Related Articles: Chemical quenching, Coincidence circuit and maximal heat dissipation through the liquid metal. The mini- for liquid scintillation counters, Dilution quenching, Pulse height mal friction allows the motor to be rotated at operational speed at analysers, Scintillator the beginning of the day and then run continuously (until stopped Further Readings: Brown, B. H. et al. 1999. Medical Physics with electromechanical brakes). and Biomedical Engineering, Institute of Physics Publishing, Related Articles: Glass envelope, Bearing, Filament heating, Bristol, UK, pp. 177–180; Graham, D. T. and P. Cloke. 2003. X-ray tube, Anode, Metal x-ray tube Principles of Radiological Physics, 4th edn., Elsevier Science Ltd., Edinburgh, UK, pp. 381–383; Knoll, G. F. 2000. Radiation Liquid scintillation (LS) counting Detection and Measurement, 3rd edn., John Wiley & Sons, (Radiation Protection) Liquid scintillation counting technique is used to measure weakly penetrating radiation like gamma PM 1 LS PM 2 Coincidence system Amplifier PHA Recording FIGURE L.44 Spiral grooves on the rotor of an x-ray tube (the sectioned device large anode is seen on the right of this model). Image taken with permis- sion from a Toshiba model of power x-ray tube with liquid metal anode bearing. (Image courtesy of Toshiba, Shizuoka-ken, Japan.) FIGURE L.45 Scheme of a liquid scintillation counting equipment. Liquid scintillators 565 Local rules Inc., New York, pp. 345–346; Saha, G. B. 2001. Physics and two ways, either by a point source or by an extended source. The Radiobiology of Nuclear Medicine, 2nd edn., Springer-Verlag, most frequently used point source is a spark gap, placed in one L New York, pp. 84–87. of the foci of an ellipsoid, immersed in water. As a high voltage is discharged over the gap, an explosive plasma formation and Liquid scintillators vaporisation of the water takes place. As all rays from the first (Nuclear Medicine) A detector principal used to estimate the foci, via the perimeter of the ellipsoid, to the second foci have activity in biological samples using liquid scintillators. equal path length, the energy will again converge in the second Samples prepared for liquid scintillation consist of at least foci, placed where the stone is located. For an extended source, three components: (1) the radioactive material being counted, (2) a large number of ultrasound transducers (many hundreds) a solvent and (3) a scintillator. Often an inert additive is added to are located in an array in the shape of a segment of a sphere. facilitate the sample preparation. The basic mechanism behind Simultaneous transmission of a high amplitude pulse on all ele- the liquid scintillator involves many phenomena. The kinetic ments will also produce a shock wave towards the centre of the energy of the emitted charged particle, typically a β-particle, sphere. A treatment using either strategy, can last up to an hour, is deposited in the solvent leaving excited scintillation atoms during which up to 10,000 shock wave pulses are emitted. The in its path. Ideally scintillation atoms emit scintillation light at stone is slowly disintegrated, and not pulverised by one single de-excitation but the scintillation process is generally accom- blow. The underlying mechanisms are not fully understood, but panied by another competing process, namely, the quenching the high pressure and rise-time are believed to be of importance, process (see separate article for more information regarding the as well as cavitation effects. quenching process). Typically a PM tube is placed close to the vial to collect and strengthen the signal induced by the scintil- Local area network (LAN) lation light. (General) A LAN is a small to medium scale computer network (as opposed to wide area networks – WANs) providing infor- List mode mation access and data sharing within a home or small office (Nuclear Medicine) One of the two modes of acquiring images setting. The LAN can link up and offer a communication route in nuclear medicine (the other being frame mode). Each of the X between computers acting as servers or just workstations, print- and Y position signals of each scintillation event received from ers, network attached storage systems and other networkable the gamma camera are stored in the computer memory in a form devices. of a data list, along with periodic timing marks. Thus, all the data The LAN usually offers much higher data transfer rates than from the camera, i.e. every recorded event, as well as the time of are available over WANs although this is dependent on the exact the event, are recorded in a ‘list’ (hence the name), forming a huge architecture and physical medium used to relay the data. The dataset. After all the data are acquired after a pre-set amount of LAN may take many physical forms including Ethernet over time, the list data are reformatted into conventional images for co-axial cable or shielded or unshielded twisted pair or Wi-Fi display. and can be

configured in several organisational forms or topolo- The advantage of list-mode data acquisition is that it allows for gies. Switches, routers or hubs can be added to a LAN to transfer enormous flexibility in data analysis and image visualisation. The information packets between separate sections of the LAN and disadvantage of this mode of data acquisition is that it requires a link the LAN to one or more other LANs or WANs usually via very large amount of computer memory and time for image pro- a modem. cessing. Most modern PET scanners offer an option of list-mode data acquisition. Local overdose (Radiotherapy) A local overdose is delivered to a patient when a higher dose is given than was intended; this can arise from accidental equipment malfunction or a miscalculation in the treatment plan or the applied monitor units. It is generally accepted that the radiation dose should be delivered to a toler- ance of within 5% of the prescribed dose with a confidence level of 95%, therefore a level of 10% as a definition of overdose would seem reasonable. However this may vary across different quality assurance systems. An overdose may increase the frequency and Related Articles: Frame mode for digital image acquisition, severity of treatment complications. Complications depend on Frame mode, Radionuclide imaging several factors such as the total delivered dose, the total duration Further Readings: Bushberg, Seibert, Leidholdt and Boone. of the treatment, the size and location of the irradiated volume 2012. The Essential Physics of Medical Imaging, 3rd edn., and the organisation of the functional subunits of the organ being Lippincott Williams & Wilkins; Cherry, Sorenson and Phelps. irradiated. 2012. Physics in Nuclear Medicine, 4th edn., Elsevier. Local rules Lithotripter (Radiation Protection) In the work with ionising radiation, (Ultrasound) A lithotripter is a device which is used for non- employers, registrants and licensees shall, in consultation with invasive treatment of kidney stones. It works by disintegrating the workers, through their representatives (if appropriate): (1) ensure stones by extracorporally induced ultrasonic shock waves. The protection and safety for workers and other persons; (2) comply shock waves form due to non-linear propagation and can have with regulations and dose limits, including investigation levels or peak positive pressures in the range of 100 MPa. authorised levels and procedures in the event that any such dose The German company Dornier Systems originally developed limit is exceeded and (3) ensure that any work is adequately super- the method. Today the shock wave can be generated in basically vised. In order to achieve safety in the procedures it is important Local rules (laser) 566 Localiser to prepare the local rules and made them known to workers and Localisation jig L any other person involved. (Radiotherapy) A localisation jig is the device with embedded The local rules should be tailored taking into account the local fiducial markers used in brachytherapy for source localisation. laws and regulations, for the various medical applications of ion- Source localisation is the determination of the 3D coordinates and ising radiation and refer, when possible, to specific equipment and orientation of each source relative to the patient anatomy. Some procedure; they should include: procedures for wearing, handling sources refer to the localisation jig as ‘reconstruction jig’. and storing personal dosimeters as well as actions to minimise The jigs are made of acrylic or plastic and contain fiducial radiation exposure during unusual events. markers in the shape of crosshairs with known sizes on each side The workers should be well informed and trained on the con- of the jig. The crosshairs are used by the planning software to tent of the local rules. In addition, a written copy should be read- create a common coordinate system. For the geometry recon- ily available in each room, where it eventually could be needed. struction, two appropriate projection images are needed. A single The persons responsible with their telephone numbers should be point in space (3D) results in two coordinates in each projection clearly indicated in case help is needed. image (2D). Known geometry of the localisation jig and the fidu- Further Reading: IAEA (International Atomic Energy cial markers enables the determination of coordinates for other Agency). 1996. International Basic Safety Standard for Protection structures projected on images (Figure L.46). against Ionizing Radiation and for the Safety of Radiation Related Articles: Brachytherapy, Afterloading Sources, Safety Series No. 115, International Atomic Energy Further Reading: Gerbaulet, A., R. Pötter, J. J. Mazeron, H. Agency, Vienna, Austria. Meertens and E. Van Limbergen, eds. 2002. The GEC ESTRO Handbook of Brachytherapy, ESTRO, Brussels, Belgium. Local rules (laser) (Non-Ionising Radiation) A local rules document should be in Localisation radiograph place for every Class 3B and Class 4 laser within an organisation (Radiotherapy) Localisation is the process of determining the and describe the normal and emergency operating procedures to tumour/target location. It is the first stage in the radiotherapy be followed during laser use. All members of staff involved in treatment planning process. This process may be achieved using laser use should have read the local rules and should be asked to tomographic imaging, such as CT or MRI or by using projection sign to say that they have done so. The MHRA guidance docu- radiographs. Localisation is often achieved by using two orthogo- ment suggests that the local rules should contain the following nal radiographs to determine the position of the tumour or an ana- information: tomical structure in three dimensions. Treatment Simulator: A treatment simulator may be used • Details of the management structure to obtain a localisation radiograph. This is a diagnostic x-ray set • Contact details for laser safety role holders (LPA, LPS, mounted on an isocentric gantry such that it has the same degree lead user) of movement as a treatment linac. The localisation radiograph • Register of authorised users may be obtained with contrast medium to locate the tumour in • Laser/IPL key keeping arrangements certain sites such as the bladder. Figure L.47 shows a treatment • Description of the controlled area simulator. • Description of the nature of the hazard associated with Abbreviations: CT = Computed tomography and MR = the equipment Magnetic resonance imaging. • Controlled and safe access to the equipment area • Training requirements for equipment users Localiser • Responsibilities of the equipment user (Magnetic Resonance) The term localiser normally refers to • Methods of safe working, including equipment layout the acquisition of three orthogonal slices of a volume of inter- • Simple pre-use safety checks and instructions est in order to facilitate pre-examination slice positioning. A • Requirements for personal protective equipment • Prevention of use by unauthorised persons • Adverse event and equipment fault procedures and logs • Use of loan or demonstration equipment • Temporary staff • Visiting engineers Related Articles: Laser classification Further Reading: Medicines and Healthcare Products Regulatory Agency, Lasers, intense light source systems and LEDs – guidance for safe use in medical, surgical, dental and aesthetic practices, Crown Copyright, September 2015. Local underdose (Radiotherapy) A local underdose consists in delivering less than the intended dose to a patient because of accidental equipment malfunction or a miscalculation in the treatment plan or applied monitor units. Underdosing may reduce the tumour control prob- ability. Many underdoses go undiscovered and may only be detected after a relatively long time and consequently may involve a large number of patients. FIGURE L.46 Localisation jig. Logic analyser 567 Longitudinal movement Longitudinal magnetisation (Magnetic Resonance) The magnetisation of tissues in the direction L of the static magnetic field B0 is designated as longitudinal magne- tisation. If a tissue is placed inside a magnetic field B0 the magnetic moments of individual protons (=spin) will begin to rotate, or pre- cess, about the magnetic field. The spins will be tilted slightly away from the axis of the magnetic field, but the axis of rotation will be parallel to B0. The tissue will therefore become magnetised in the presence of B0 with a value M0 known as the net magnetisation (i.e. vector sum of all spins). Longitudinal magnetisation cannot, however, directly produce a RF signal. If a radiofrequency pulse is applied along the x-axis it bends the magnetisation M0 away from the z-axis and causes it to precess around a new direction (Beff). A RF pulse, indicated as 90° pulse and creating an additional mag- netic field B1, with a central frequency ω0 and orientation perpen- dicular to B0 resulting in an effective field Beff = B0 + B1 will cause M0 to rotate entirely into the transverse plane (Figure L.49). There will be no longitudinal magnetisation following the 90° pulse. The longitudinal magnetisation along the z-axis has been con- verted into a detectable magnetisation in the x, y plane resulting in a continuously varying magnetic field which induces an alternating current in an external receiver coil. After the 90° pulse the MR sig- FIGURE L.47 Radiotherapy treatment simulator. nal declines due to relaxation processes and field inhomogeneities. This return of magnetisation follows an exponential growth process with T1 being the time constant for the growth localiser scan commonly employs a T1-weighted gradient-echo (Figure L.50). After three T1 time periods M will return to 95% of pulse sequence and is also referred to as ‘Scout’ or ‘Survey’ its value prior to the excitation. depending on the MRI unit vendor. Related Articles: Gradient echo (GE), T1-weighted Longitudinal movement (Nuclear Medicine) This is movement along the long axis (feet to Logic analyser head) of the body. Patients in scanners or emission cameras can (General) A logic analyser is an electronic instrument that mea- sures and displays signals in digital circuits. Logic analysers are used for capturing data in systems that have many digital channels z z and therefore cannot be displayed with an oscilloscope. It usually B0 y y has an embedded computer which runs software enabling conver- 90° sion of the measured data into a form more appropriate for analy- M pulse sis: timing diagrams, assembly language, state machine traces, or x x others (Figure L.48). M Hyperlink: http: / /en. wikip edia. org /w iki /L ogic_ analy zer B1 FIGURE L.49 Effect of a 90° RF pulse in rotating frame of reference. 1–exp(–τ/T1) 1.0 0.95 0.86 0.63 Mz M0 0 T1 2 * T1 3 * T1 FIGURE L.50 T1 relaxation curve. Following a 90° RF pulse there is no FIGURE L.48 Logic analyser. longitudinal magnetisation. Longitudinal travel 568 Lossless compression induce image artefacts originating from patient movement during L the acquisition. Motion artefacts can be difficult to correct for and may lead to misdiagnosis if not properly attended to. It is there- fore important to minimise patient movement during imaging. Longitudinal travel (Nuclear Medicine) See Longitudinal movement Longitudinal wave (Ultrasound) When an ultrasound wave propagates in a medium, the particles within the medium will start oscillating. In gases, liquids and soft tissue this oscillation is always in the same direction as the ultrasound wave, giving rise to longitudinal (or compressional) wave propagation, Figure L.51. The particles’ dis- placement and velocity depend on the medium’s acoustic imped- ance, the pressure amplitude and the frequency of the ultrasound wave (see Displacement). Related Articles: Displacement, Compression, Transversal wave, Lamb wave Long-term morbidity (Radiotherapy) Long-term morbidity relates to effects due to radiotherapy that persist for a relatively long period after the treat- ment. A late radiation morbidity scoring schema has been issued by RTOG/EORTC (Figure L.52). Further Reading: Cox J.D. et al. 1995. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys. Mar 30;31(5):1341–6. Related Articles: Adverse effects (Radiotherapy), Late response of normal tissue, Tolerance Lookup table (Diagnostic Radiology) Lookup table (look-up table, LUT) is an array of data used by the computer to convert data from one form to another. LUTs are widely used in image processing, for example for storing palettes of colour (e.g. the 8 bits in a byte used to address 256 cells each holding information about a particular colour). A typical use of LUTs in medical imaging is for the con- version of the pixel numbers into physical colours (or shades of grey). Lorentzian lineshape (Nuclear Medicine) The Lorentzian lineshape refers to shape of a Lorentzian distribution, also known as a Cauchy distribution. The Lorentzian distribution has a similar bell shape as