Upload 26 files

.gitattributes

@@ -57,3 +57,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+data/generate_data/test_balanced.json filter=lfs diff=lfs merge=lfs -text
+data/generate_data/test.json filter=lfs diff=lfs merge=lfs -text
+data/generate_data/train.json filter=lfs diff=lfs merge=lfs -text
+data/generate_data/val.json filter=lfs diff=lfs merge=lfs -text
+data/masks_aa filter=lfs diff=lfs merge=lfs -text
+data/masks_ab filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,71 @@
+# 3D ReasonKnee: 3D Grouned Reasoning Dataset
+
+This repository contains code for preprocessing, generating, and analyzing 3D knee MRI data for osteoarthritis assessment.
+
+## Repository Structure
+
+### Data Organization
+
+- `/data/`
+  - `/OAI/` - Raw DICOM files from OAI dataset
+  - `/labels/` - Contains taxonomymatch csv files with MOAKS scores
+  - `/images/` - Contains the preprocessed images
+  - `/masks.zip` - Contains the segmentation masks of all subregions and lesions given by nnU-Net, please use `cat masks_aa masks_ab > masks.zip` and `unzip masks.zip`
+  - `/split_files/` - Contains patient_id,time,study_folder,series_folder,knee_side,path,image_filepath,label_filepath infomation of all studies with random train/val/test split files
+  - `/question_files/` - Contains mapping files for subregions and grades
+    -  `questions.json` - Queries created based on the MOAKS framework 
+    -  `cot_question_mapping.json` - Expert Chain of Thought of each question 
+    -  `labels.json` - label subregion/lesion mapping of the masks
+    -  `grade_dict.json` - Example structure of all possible scores
+    -  `question_subregion_mapping.json` - Corresponding subregion of each question
+  - `/generate_data/` - Contains final generated data with image paths (image), prompts (prompt), chain-of-thought reasoning (cot), and final diagnoses including bounding boxes of subregions and grade assessments (final_diagnosis).
+
+### Code Organization
+
+- `/code/`
+  - `/data_preprocessing/` - Scripts for initial data preparation of raw DICOM files
+    - `dcm2nifti.py` - Converts DICOM files to NIfTI format and save to `./data/images/`
+
+## How to Use 
+
+1. **OAI Download**: Please download the original raw DICOM files from https://oai.ucsf.edu/oai-osteoarthritis-initiative
+2. **DICOM to NIfTI Conversion**: Use `/code/data_preprocessing/dcm2nifti.py` to convert DICOM files to NIfTI format 
+3. **Data Example**: data file can be found in `/generate_data/`
+
+## Example Data Format
+
+Below is an example of the data format in our generated data files:
+```json
+  {
+    "image": "/data/images/9021195/48M/20090518_12801512_1.nii.gz",
+    "prompt": "### Question:\nIn this DESS MRI, can you diagnose cartilage lesions in the femur medial central subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur medial central subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"final_diagnosis\": {\"femur medial central cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}\n### Answer (think step by step):\n",
+    "answer": "The cartilage looks gray in the DESS image. It covers the articular surfaces of the bones. I need to see whether the patient's cartilage has a smooth surface, evenly distributed thickness. If there is an abrupt thinning of cartilage, disruption of contour, it implies a cartilage lesion. There is no abrupt change in cartilage contour or any absence of cartilage in this subregion.\n Final diagnosis: {\"bounding_box\": [83, 141, 117, 261, 154, 258], \"final_diagnosis\": {\"femur medial central cartilage lesion\": {\"size\": 0, \"depth\": 0}}}",
+    "cot": "The cartilage looks gray in the DESS image. It covers the articular surfaces of the bones. I need to see whether the patient's cartilage has a smooth surface, evenly distributed thickness. If there is an abrupt thinning of cartilage, disruption of contour, it implies a cartilage lesion. There is no abrupt change in cartilage contour or any absence of cartilage in this subregion.",
+    "final_diagnosis": {
+      "bounding_box": [
+        83,
+        141,
+        117,
+        261,
+        154,
+        258
+      ],
+      "final_diagnosis": {
+        "femur medial central cartilage lesion": {
+          "size": 0,
+          "depth": 0
+        }
+      }
+    }
+  }
+```
+
+### License and Access Restrictions
+The Osteoarthritis Initiative (OAI) is a multi-center, ten-year observational study of men and women, sponsored by the National Institutes of Health (part of the Department of Health and Human Services). To obtain access to OAI data you will need an NDA user account. Account can be requested directly from the NIMH Data Archive (NDA) website https://nda.nih.gov/oai/query-download.
+
+Our 3D-GROI dataset is released under the [Creative Commons Attribution 4.0 International (CC BY 4.0)](https://creativecommons.org/licenses/by/4.0/) license. This allows for:
+- Sharing and adaptation of the material
+- Commercial use
+- With the requirement to provide appropriate attribution
+
+

code/data_preprocessing/dcm2nifti.py

@@ -0,0 +1,83 @@
+import os
+import pydicom
+import nibabel as nib
+import pandas as pd
+import dicom2nifti
+from tqdm import tqdm
+import concurrent.futures
+
+def get_knee_side(dicom_path):
+    """
+    Reads DICOM metadata and extracts the knee side from SeriesDescription.
+    Returns:
+        int: 1 for RIGHT, 2 for LEFT (based on SeriesDescription), or None on error.
+    """
+    try:
+        ds = pydicom.dcmread(dicom_path, stop_before_pixels=True)
+        series_desc = getattr(ds, "SeriesDescription", "").upper()
+        print(series_desc)
+        return 'RIGHT' if series_desc == "SAG_3D_DESS_RIGHT" else 'LEFT'
+    except Exception as e:
+        print(f"Error reading DICOM file {dicom_path}: {e}")
+        return None
+
+def convert_dcm_to_nifti(dicom_path, save_image_filepath):
+    """
+    Converts DICOM files to NIfTI format.
+    Args:
+        dicom_path (str): Path to the DICOM file.
+        save_image_filepath (str): Path to save the NIfTI file.
+    """
+    dicom2nifti.dicom_series_to_nifti(dicom_path, save_image_filepath, reorient_nifti=True)
+    print(f"Saved {save_image_filepath}")
+    return save_image_filepath
+
+def process_row(row, save_image_folder):
+    """
+    Process a single row from the DataFrame.
+    """
+    patient_id = row['patient_id']
+    time = row['time']
+    study_folder = row['study_folder']
+    series_folder = row['series_folder']
+    knee_side = row['knee_side']
+    path = row['path']
+    
+    save_image_filepath = os.path.join(save_image_folder, str(patient_id), str(time), 
+                                       f'{study_folder}_{series_folder}_{knee_side}.nii.gz')
+    
+    if os.path.exists(save_image_filepath):
+        print(f"Skipping {save_image_filepath} because it already exists")
+    else:
+        os.makedirs(os.path.dirname(save_image_filepath), exist_ok=True)
+        convert_dcm_to_nifti(path, save_image_filepath)
+    
+    row_copy = row.copy()
+    row_copy['save_image_filepath'] = save_image_filepath
+    return row_copy
+
+if __name__ == "__main__":
+    save_image_folder = '/data/images'
+    df = pd.read_csv('/data/all_studies.csv')
+    
+    # Define number of workers (threads)
+    max_workers = 8  # You can adjust this based on your CPU cores
+    save_rows = []
+    # Process in batches if memory is a concern
+    batch_size = 100
+    for i in range(0, len(df), batch_size):
+        batch_df = df.iloc[i:i+batch_size]
+        
+        with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
+            futures = [executor.submit(process_row, row, save_image_folder) 
+                    for _, row in batch_df.iterrows()]
+            
+            for future in tqdm(concurrent.futures.as_completed(futures), 
+                            total=len(futures), 
+                            desc=f"Batch {i//batch_size+1}/{(len(df)+batch_size-1)//batch_size}"):
+                try:
+                    result = future.result()
+                    save_rows.append(result)
+                except Exception as e:
+                    print(f"Processing error: {e}")
+    

data/generate_data/test.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95bf3620082d849ba8b17cff3c9e22c9652f3a3d3d92e479a004740b2afaf7de
+size 171712528

data/generate_data/test_balanced.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9539a4e98e207965f3bb2dbd03d891ae2363abfbfb25a0ba61c2e567f2a97e0f
+size 40626553

data/generate_data/train.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c9141aba4c2c48288f03c2ead8e5c11afae507bc32adeeb6fdf21237f56e5ae
+size 837738466

data/generate_data/val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50668bdc31419fe1084cce07fea37554a5f874798b94c63d54f3f291a0df52ca
+size 116376956

data/labels/manual_annotated_kMRI_18M.csv

@@ -0,0 +1,7 @@
+ID,SIDE,VERSION,,READPRJ,V00MCMPM,V00MCMPM.1,V00MCMPL,V00MCMPL.1,V00MCMFMA,V00MCMFMA.1,V00MCMFLA,V00MCMFLA.1,V00MCMFMP,V00MCMFMP.1,V00MCMFLP,V00MCMFLP.1,V00MCMFMC,V00MCMFMC.1,V00MCMFLC,V00MCMFLC.1,V00MCMTMA,V00MCMTMA.1,V00MCMTLA,V00MCMTLA.1,V00MCMTMC,V00MCMTMC.1,V00MCMTLC,V00MCMTLC.1,V00MCMTMP,V00MCMTMP.1,V00MCMTLP,V00MCMTLP.1,V00MBMSFMA,V00MBMPFMA,V00MBMNFMA,V00MBMSFLA,V00MBMPFLA,V00MBMNFLA,V00MBMSFMC,V00MBMPFMC,V00MBMNFMC,V00MBMSFLC,V00MBMPFLC,V00MBMNFLC,V00MBMSFMP,V00MBMPFMP,V00MBMNFMP,V00MBMSFLP,V00MBMPFLP,V00MBMNFLP,V00MBMSSS,V00MBMPSS,V00MBMNSS,V00MBMSTMA,V00MBMPTMA,V00MBMNTMA,V00MBMSTLA,V00MBMPTLA,V00MBMNTLA,V00MBMSTMC,V00MBMPTMC,V00MBMNTMC,V00MBMSTLC,V00MBMPTLC,V00MBMNTLC,V00MBMSTMP,V00MBMPTMP,V00MBMNTMP,V00MBMSTLP,V00MBMPTLP,V00MBMNTLP,V00MBMSPM,V00MBMPPM,V00MBMNPM,V00MBMSPL,V00MBMPPL,V00MBMNPL,V00MMTMA,V00MMTLA,V00MMTMB,V00MMTLB,V00MMTMP,V00MMTLP,V00MMHMA,V00MMXMM,V00MMXMA,V00MMXLA,V00MMXLL,V00MMRTM,V00MOSPS,V00MOSPI,V00MOSPM,V00MOSPL,V00MOSFMA,V00MOSFLA,V00MOSFMP,V00MOSFLP,V00MOSFMC,V00MOSFLC,V00MOSTM,V00MOSTL,V00MACLTR,V00MACLBML+V00MGCACL,V00MACLRP,V00MPCLTR,V00MPCLBML+V00MGCPCL,V00MPTSIG,V00MGCTIB+V00MGCSM+V00MGSST+V00MGCOTH,V00MANSBUR,V00MIPBUR,V00MPPBUR,V00MSYIC,V00MEFFWK,V00MITBSIG,V00MPOPCYS,,,
+9007827,1,,999,0,0,0,0,2,2,0,0,0,0,2,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,3,1,1,1,1,0,0,0,1,1,2,1,0,1,0,0,0,0,0,0,0,0,1,0,0,0
+9016304,1,,999,0,0,0,0,2,2,0,0,0,0,2,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,2,1,0,0,0,0,1,0,0,0,0,0,0,0,3,1,1,1,1,0,0,0,1,1,2,1,0,0,0,0,0,0,0,0,0,0,1,3,0,0
+9028786,1,,999,0,0,0,0,2,2,0,0,0,0,2,2,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,2,1,0,0,0,0,1,0,0,0,0,0,0,0,1,1,1,1,1,0,0,0,1,1,1,1,0,0,0,0,0,1,0,0,0,0,0,3,0,0
+9021791,1,,999,0,0,0,0,2,2,0,0,0,0,2,2,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,2,1,0,0,0,0,1,0,0,0,0,0,0,0,2,1,1,1,1,0,0,0,1,1,1,1,0,0,0,0,0,1,0,0,0,0,0,0,0,0
+9019287,1,,999,0,0,0,0,1,3,0,0,0,0,2,2,1,1,0,0,1,1,0,0,0,0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,2,1,0,0,0,0,1,0,0,0,0,0,0,0,2,1,1,1,1,0,0,0,1,1,1,1,0,0,0,0,0,1,0,0,0,0,0,1,0,0
+9000798,1,,999,0,0,0,0,1,3,0,0,0,0,2,2,2,2,0,0,1,2,0,0,2,2,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,2,1,1,1,1,0,0,0,0,3,0,0,0,0,0,0,0,2,1,1,1,1,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,2,0,0

data/masks_aa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:970fd9b1da59b026acf925ae03f8b7a44b07361172fb02f0a733a7cb3bb052c4
+size 2147483648

data/masks_ab

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f52c75690c585659fa66153c84ea63876866ffb1ba942523b54dc67cc3ac5580
+size 873532299

data/question_files/grade_dict.json

@@ -0,0 +1,241 @@
+{
+    "patella medial cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "patella lateral cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur medial trochlea cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur lateral trochlea cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur medial posterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur lateral posterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur medial central cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur lateral central cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia medial anterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia lateral anterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia medial central cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia lateral central cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia medial posterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "tibia lateral posterior cartilage lesion":{
+        "size": 0,
+        "depth": 0
+    },
+    "femur medial trochlea BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "femur lateral trochlea BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "femur medial central BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "femur lateral central BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "femur medial posterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "femur lateral posterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia subspinous BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia medial anterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia lateral anterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia medial central BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia lateral central BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia medial posterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "tibia lateral posterior BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "patella medial BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "patella lateral BML":{
+        "Size of BML (including cyst)": 0,
+        "Number of BML within subregion": 0,
+        "% of lesion that is BML (excluding cyst)": 0
+    },
+    "Medial meniscus anterior morphology":{
+        "morphology": 0
+    },
+    "Lateral meniscus anterior morphology":{
+        "morphology": 0
+    },
+    "Medial meniscus body morphology":{
+        "morphology": 0
+    },
+    "Lateral meniscus body morphology":{
+        "morphology": 0
+    },
+    "Medial meniscus posterior morphology":{
+        "morphology": 0
+    },
+    "Lateral meniscus posterior morphology":{
+        "morphology": 0
+    },
+    "Meniscal hypertrophy":{
+        "location": 0
+    },
+    "Medial meniscus medial extrusion":{
+        "grade": 0
+    },
+    "Medial meniscus anterior extrusion":{
+        "grade": 0
+    },
+    "Lateral meniscus anterior extrusion":{
+        "grade": 0
+    },
+    "Lateral meniscus lateral extrusion":{
+        "grade": 0
+    },
+    "Meniscal root tear":{
+        "location": 0
+    },
+    "patella superior osteophyte":{
+        "grade": 0
+    },
+    "patella inferior osteophyte":{
+        "grade": 0
+    },
+    "patella medial osteophyte":{
+        "grade": 0
+    },
+    "patella lateral osteophyte":{
+        "grade": 0
+    },
+    "femur medial anterior osteophyte":{
+        "grade": 0
+    },
+    "femur lateral anterior osteophyte":{
+        "grade": 0
+    },
+    "femur medial posterior osteophyte":{
+        "grade": 0
+    },
+    "femur lateral posterior osteophyte":{
+        "grade": 0
+    },
+    "femur medial central osteophyte":{
+        "grade": 0
+    },
+    "femur lateral central osteophyte":{
+        "grade": 0
+    },
+    "tibia medial osteophyte":{
+        "grade": 0
+    },
+    "tibia lateral osteophyte":{
+        "grade": 0
+    },
+    "ACL":{
+        "Complete Tear?": 0,
+        "Associated with BML/cyst at site of insertion or origin?": 0,
+        "ACL Repair": 0
+    },
+    "PCL":{
+        "Complete Tear?": 0,
+        "Associated with BML/cyst at site of insertion or origin?": 0
+    },
+    "Patellar Tendon":{
+        "Signal abnormality": 0
+    },
+    "Ganglion cyst":{
+        "grade": 0
+    },
+    "Pes anserine bursitis":{
+        "grade": 0
+    },
+    "Infrapatellar bursitis":{
+        "grade": 0
+    },
+    "Prepatellar bursitis":{
+        "grade": 0
+    },
+    "Hoffa-Synovitis":{
+        "grade": 0
+    },
+    "Effusion-Synovitis":{
+        "grade": 0
+    },
+    "Iliotibial band signal":{
+        "grade": 0
+    },
+    "Popliteal cyst":{
+        "grade": 0
+    }
+}

data/question_files/labels.json

@@ -0,0 +1,67 @@
+{
+    "background": 0,
+    "effusion-synovitis": 1,
+    "femur lateral anterior osteophyte": 2,
+    "femur lateral central bml": 3,
+    "femur lateral central cartilage lesion": 4,
+    "femur lateral central osteophyte": 5,
+    "femur lateral posterior bml": 6,
+    "femur lateral posterior cartilage lesion": 7,
+    "femur lateral posterior osteophyte": 8,
+    "femur lateral trochlea bml": 9,
+    "femur lateral trochlea cartilage lesion": 10,
+    "femur medial anterior osteophyte": 11,
+    "femur medial central bml": 12,
+    "femur medial central cartilage lesion": 13,
+    "femur medial central osteophyte": 14,
+    "femur medial posterior bml": 15,
+    "femur medial posterior cartilage lesion": 16,
+    "femur medial posterior osteophyte": 17,
+    "femur medial trochlea bml": 18,
+    "femur medial trochlea cartilage lesion": 19,
+    "hoffa-synovitis": 20,
+    "patella inferior osteophyte": 21,
+    "patella lateral bml": 22,
+    "patella lateral cartilage lesion": 23,
+    "patella lateral osteophyte": 24,
+    "patella medial bml": 25,
+    "patella medial cartilage lesion": 26,
+    "patella medial osteophyte": 27,
+    "patella superior osteophyte": 28,
+    "tibia lateral central bml": 29,
+    "tibia lateral central cartilage lesion": 30,
+    "tibia lateral osteophyte": 31,
+    "tibia lateral posterior bml": 32,
+    "tibia medial anterior bml": 33,
+    "tibia medial anterior cartilage lesion": 34,
+    "tibia medial central bml": 35,
+    "tibia medial central cartilage lesion": 36,
+    "tibia medial osteophyte": 37,
+    "tibia medial posterior bml": 38,
+    "tibia subspinous bml": 39,
+    "acl": 40,
+    "lateral meniscus": 41,
+    "lateral meniscus anterior morphology": 42,
+    "lateral meniscus posterior morphology": 43,
+    "medial meniscus": 44,
+    "medial meniscus anterior morphology": 45,
+    "medial meniscus posterior morphology": 46,
+    "patellar tendon": 47,
+    "pcl": 48,
+    "tibia cartilage": 49,
+    "lateral_central_femur": 50,
+    "lateral_posterior_femur": 51,
+    "lateral_trochlea_femur": 52,
+    "medial_central_femur": 53,
+    "medial_posterior_femur": 54,
+    "medial_trochlea_femur": 55,
+    "patella_lateral": 56,
+    "patella_medial": 57,
+    "tibia_lateral_anterior": 58,
+    "tibia_lateral_central": 59,
+    "tibia_lateral_posterior": 60,
+    "tibia_medial_anterior": 61,
+    "tibia_medial_central": 62,
+    "tibia_medial_posterior": 63,
+    "tibia_subspinous": 64
+}

data/question_files/question_subregion_mapping.json

@@ -0,0 +1,71 @@
+{
+    "1":  "patella_medial",
+    "2":  "patella_lateral",
+    "3":  "medial_trochlea_femur",
+    "4":  "medial_central_femur",
+    "5":  "medial_posterior_femur",
+    "6":  "lateral_trochlea_femur",
+    "7":  "lateral_central_femur",
+    "8":  "lateral_posterior_femur",
+    "9":  "tibia_medial_anterior",
+    "10": "tibia_medial_central",
+    "11": "tibia_medial_posterior",
+    "12": "tibia_lateral_anterior",
+    "13": "tibia_lateral_central",
+    "14": "tibia_lateral_posterior",
+    "15": "tibia_subspinous",
+    "16": "patella_medial",
+    "17": "patella_lateral",
+    "18": "medial_trochlea_femur",
+    "19": "medial_central_femur",
+    "20": "medial_posterior_femur",
+    "21": "lateral_trochlea_femur",
+    "22": "lateral_central_femur",
+    "23": "lateral_posterior_femur",
+    "24": "tibia_medial_anterior",
+    "25": "tibia_medial_central",
+    "26": "tibia_medial_posterior",
+    "27": "tibia_lateral_anterior",
+    "28": "tibia_lateral_central",
+    "29": "tibia_lateral_posterior",
+    "30": ["patella_medial", "patella_lateral"],
+    "31": ["patella_medial", "patella_lateral"],
+    "32": "patella_medial",
+    "33": "patella_lateral",
+    "34": "medial_trochlea_femur",
+    "35": "lateral_trochlea_femur",
+    "36": "medial_posterior_femur",
+    "37": "lateral_posterior_femur",
+    "38": "lateral_central_femur",
+    "39": "lateral_posterior_femur",
+    "40": "tibia_medial_anterior",
+    "41": "tibia_lateral_anterior",
+    "42": "hoffa-synovitis",
+    "43": "effusion-synovitis",
+    "44": "medial meniscus",
+    "45": "medial meniscus",
+    "46": "lateral meniscus",
+    "47": "lateral meniscus",
+    "48": "medial meniscus",
+    "49": "medial meniscus",
+    "50": "medial meniscus",
+    "51": "lateral meniscus",
+    "52": "lateral meniscus",
+    "53": "lateral meniscus",
+    "54": "medial meniscus", 
+    "55": "lateral meniscus",
+    "56": "medial meniscus",
+    "57": "lateral meniscus",
+    "58": "medial meniscus",
+    "59": "lateral meniscus",
+    "60": "acl",
+    "61": "pcl",
+    "62": "patellar tendon",
+    "63": "",
+    "64": "",
+    "65": "",
+    "66": "",
+    "67": "",
+    "68": "",
+    "69": ""  
+}

data/question_files/questions.json

@@ -0,0 +1,71 @@
+{
+  "1": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the patella medial subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the patella medial subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella medial BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "2": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the patella lateral subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the patella lateral subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella lateral BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "3": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur medial trochlea subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur medial trochlea subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial trochlea BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "4": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur medial central subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur medial central subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial central BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "5": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur medial posterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur medial posterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial posterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "6": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur lateral trochlea subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur lateral trochlea subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral trochlea BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "7": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur lateral central subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur lateral central subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral central BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "8": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the femur lateral posterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the femur lateral posterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral posterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "9": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia medial anterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia medial anterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial anterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "10": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia medial central subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia medial central subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial central BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "11": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia medial posterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia medial posterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial posterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "12": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia lateral anterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia lateral anterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral anterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "13": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia lateral central subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia lateral central subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral central BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "14": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia lateral posterior subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia lateral posterior subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral posterior BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "15": "In this DESS MRI, can you diagnose bone-marrow lesions (BMLs) in the tibia subspinous subregion? BMLs are scored using three MOAKS components: Size (Grade 0: None, Grade 1: <33% of subregion, Grade 2: 33-66% of subregion, Grade 3: >66% of subregion), Number (Grade 0: None, Grade 1: 1 BML, Grade 2: 2 BMLs, Grade 3: ≥3 BMLs), and % of lesion that is BML excluding cyst (Grade 0: None, Grade 1: <33%, Grade 2: 33-66%, Grade 3: >66%). Provide your reasoning, the bounding box of the tibia subspinous subregion, and the MOAKS scores for size, number of BMLs, and percent of lesion that is BML. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia subspinous BML\": {\"Size of BML (including cyst)\": (grade from 0-3), \"Number of BML within subregion\": (grade from 0-3), \"% of lesion that is BML (excluding cyst)\": (grade from 0-3)}}}",
+  "16": "In this DESS MRI, can you diagnose cartilage lesions in the patella medial subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the patella medial subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella medial cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "17": "In this DESS MRI, can you diagnose cartilage lesions in the patella lateral subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the patella lateral subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella lateral cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "18": "In this DESS MRI, can you diagnose cartilage lesions in the femur medial trochlea subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur medial trochlea subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial trochlea cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "19": "In this DESS MRI, can you diagnose cartilage lesions in the femur medial central subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur medial central subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial central cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "20": "In this DESS MRI, can you diagnose cartilage lesions in the femur medial posterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur medial posterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial posterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "21": "In this DESS MRI, can you diagnose cartilage lesions in the femur lateral trochlea subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur lateral trochlea subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral trochlea cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "22": "In this DESS MRI, can you diagnose cartilage lesions in the femur lateral central subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur lateral central subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral central cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "23": "In this DESS MRI, can you diagnose cartilage lesions in the femur lateral posterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the femur lateral posterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral posterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "24": "In this DESS MRI, can you diagnose cartilage lesions in the tibia medial anterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia medial anterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial anterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "25": "In this DESS MRI, can you diagnose cartilage lesions in the tibia medial central subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia medial central subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial central cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "26": "In this DESS MRI, can you diagnose cartilage lesions in the tibia medial posterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia medial posterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial posterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "27": "In this DESS MRI, can you diagnose cartilage lesions in the tibia lateral anterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia lateral anterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral anterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "28": "In this DESS MRI, can you diagnose cartilage lesions in the tibia lateral central subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia lateral central subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral central cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "29": "In this DESS MRI, can you diagnose cartilage lesions in the tibia lateral posterior subregion? Cartilage lesions are scored using two MOAKS components: Size (Grade 0: None, Grade 1: <10% of surface area, Grade 2: 10-75% of surface area, Grade 3: >75% of surface area) and Percent of full-thickness loss (Grade 0: No full-thickness loss, Grade 1: <10% of cartilage defect shows full-thickness loss, Grade 2: 10-75% of cartilage defect shows full-thickness loss, Grade 3: >75% of cartilage defect shows full-thickness loss). Provide your reasoning, the bounding box of the tibia lateral posterior subregion, and the MOAKS scores for size and depth. The output format should be a JSON object with the bounding box coordinates and the MOAKS scores: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral posterior cartilage lesion\": {\"Size of lesion\": (grade from 0-3), \"Depth of lesion\": (grade from 0-3)}}}",
+  "30": "In this DESS MRI, identify and grade the osteophyte at the patella superior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the patella region, and the MOAKS grade. Represent the MOAKS grade in this format: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella superior osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "31": "In this DESS MRI, identify and grade the osteophyte at the patella inferior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the patella region, and the MOAKS grade.  Represent the MOAKS grade in this format: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella inferior osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "32": "In this DESS MRI, identify and grade the osteophyte at the patella medial location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the patella medial region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"patella medial osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "33": "In this DESS MRI, identify and grade the osteophyte at the patella lateral location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the patella lateral region, and the MOAKS grade.  Represent the MOAKS grade in this format: {\"patella lateral osteophyte\": {\"grade\": (grade from 0-3)}}",
+  "34": "In this DESS MRI, identify and grade the osteophyte at the femur medial anterior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur medial anterior region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial anterior osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "35": "In this DESS MRI, identify and grade the osteophyte at the femur medial central location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur medial central region, and the MOAKS grade.  Represent the MOAKS grade in this format: {\"femur medial central osteophyte\": {\"grade\": (grade from 0-3)}}",
+  "36": "In this DESS MRI, identify and grade the osteophyte at the femur medial posterior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur medial posterior region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur medial posterior osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "37": "In this DESS MRI, identify and grade the osteophyte at the femur lateral anterior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur lateral anterior region, and the MOAKS grade.  Represent the MOAKS grade in this format: {\"femur lateral anterior osteophyte\": {\"grade\": (grade from 0-3)}}",
+  "38": "In this DESS MRI, identify and grade the osteophyte at the femur lateral central location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur lateral central region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral central osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "39": "In this DESS MRI, identify and grade the osteophyte at the femur lateral posterior location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the femur lateral posterior region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"femur lateral posterior osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "40": "In this DESS MRI, identify and grade the osteophyte at the tibia medial location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the tibia medial region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia medial osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "41": "In this DESS MRI, identify and grade the osteophyte at the tibia lateral location. Osteophytes are scored using MOAKS grading: Grade 0: None, Grade 1: Small, Grade 2: Medium, Grade 3: Large. If there are multiple osteophytes in a given subregion, the largest will be evaluated. The size of osteophyte should reflect protuberance (how far the osteophyte extends from the joint) rather than total volume of osteophyte. Provide your reasoning, the bounding box of the tibia lateral region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"tibia lateral osteophyte\": {\"grade\": (grade from 0-3)}}}",
+  "42": "In this DESS MRI, assess Hoffa-synovitis. Hoffa-synovitis is scored using MOAKS grading based on the hyperintensity (brightness) in Hoffa's fat pad: Grade 0: Normal (fat pad looks dark throughout with minimal traceable linear streaks), Grade 1: Mild (fat pad looks dark in general but with some mild areas of hyperintensity), Grade 2: Moderate (fat pad shows some moderate areas of hyperintensity), Grade 3: Severe (fat pad shows severe areas of hyperintensity). Provide your reasoning, the bounding box of the Hoffa region, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS score: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"Hoffa-Synovitis\": {\"grade\": (grade from 0-3)}}}",
+  "43": "In this DESS MRI, assess effusion-synovitis. Effusion-synovitis is scored using MOAKS grading based on the size of the fluid signal in the joint cavity: Grade 0: Physiologic amount of fluid, Grade 1: Small amount of fluid continuous in the retropatellar space, Grade 2: Medium amount of fluid, with slight convexity of the suprapatellar bursa, Grade 3: Large amount of fluid, causing capsular distension. Provide your reasoning, the bounding box of the effusion, and the MOAKS grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"effusion-synovitis\": {\"grade\": (grade from 0-3)}}}",
+  "44": "In this DESS MRI, measure medial meniscus anterior extrusions. Meniscal extrusion is measured as the maximum protrusion distance from the tibial plateau. Meniscal extrusion is scored using MOAKS grading: Grade 0: Extrusion is below 2mm, Grade 1: Extrusion is between 2 to 3mm, Grade 2: Extrusion is between 3 to 4.9mm, Grade 3: Extrusion is over 5mm. Provide your reasoning, the bounding box of the medial meniscus region, and the MOAKS extrusion grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"medial meniscus anterior extrusion\": (grade from 0-3)}}",
+  "45": "In this DESS MRI, measure medial meniscus medial extrusions. Meniscal extrusion is measured as the maximum protrusion distance from the tibial plateau. Meniscal extrusion is scored using MOAKS grading: Grade 0: Extrusion is below 2mm, Grade 1: Extrusion is between 2 to 3mm, Grade 2: Extrusion is between 3 to 4.9mm, Grade 3: Extrusion is over 5mm. Provide your reasoning, the bounding box of the medial meniscus region, and the MOAKS extrusion grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_grade\": {\"medial meniscus medial extrusion\": (grade from 0-3)}}",
+  "46": "In this DESS MRI, measure lateral meniscus anterior extrusions. Meniscal extrusion is measured as the maximum protrusion distance from the tibial plateau. Meniscal extrusion is scored using MOAKS grading: Grade 0: Extrusion is below 2mm, Grade 1: Extrusion is between 2 to 3mm, Grade 2: Extrusion is between 3 to 4.9mm, Grade 3: Extrusion is over 5mm. Provide your reasoning, the bounding box of the lateral meniscus region, and the MOAKS extrusion grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"lateral meniscus anterior extrusion\": {\"grade\": (grade from 0-3)}}}",
+  "47": "In this DESS MRI, measure lateral meniscus lateral extrusions. Meniscal extrusion is measured as the maximum protrusion distance from the tibial plateau. Meniscal extrusion is scored using MOAKS grading: Grade 0: Extrusion is below 2mm, Grade 1: Extrusion is between 2 to 3mm, Grade 2: Extrusion is between 3 to 4.9mm, Grade 3: Extrusion is over 5mm. Provide your reasoning, the bounding box of the lateral meniscus region, and the MOAKS extrusion grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"lateral meniscus lateral extrusion\": {\"grade\": (grade from 0-3)}}}",
+  "48": "In this DESS MRI, describe the medial meniscus anterior morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the medial meniscus region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"medial meniscus anterior morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "49": "In this DESS MRI, describe the medial meniscus body morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the medial meniscus region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"medial meniscus body morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "50": "In this DESS MRI, describe the medial meniscus posterior morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the medial meniscus region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"medial meniscus posterior morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "51": "In this DESS MRI, describe the lateral meniscus anterior morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the lateral meniscus region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_grade\": {\"lateral meniscus anterior morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "52": "In this DESS MRI, describe the lateral meniscus body morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the lateral meniscus body region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_grade\": {\"lateral meniscus body morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "53": "In this DESS MRI, describe the lateral meniscus posterior morphology. Meniscal morphology is scored using MOAKS grading: Grade 0: Normal meniscus, Grade 1: Signal abnormality (not severe enough to be considered a tear), Grade 2: Radial tear, Grade 3: Horizontal tear, Grade 4: Vertical tear, Grade 5: Complex tear, Grade 6: Partial maceration, Grade 7: Progressive partial maceration (only for follow-up), Grade 8: Complete maceration. Provide your reasoning, the bounding box of the lateral meniscus region, and the MOAKS morphology grade. The output format should be a JSON object with the bounding box coordinates and the MOAKS grade: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"MOAKS_scores\": {\"lateral meniscus posterior morphology\": {\"morphology\": (grade from 0-8)}}}",
+  "54": "In this DESS MRI, detect a medial meniscal cyst, if present. Provide your reasoning, the bounding box of the medial meniscus, and the location of any tear. Tear location is encoded as one of the following integers: \n0 = MMA (Medial Meniscus Anterior), \n1 = MMB (Medial Meniscus Body), \n2 = MMP (Medial Meniscus Posterior), 3 = No tear detected. The output format should be a JSON object with the bounding box coordinates and the cyst presence: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"medial_meniscal_cyst\": {\"presence\": 0 | 1, \"tear_location\": (0-3)}}.",
+  "55": "In this DESS MRI, detect a lateral meniscal cyst, if present. Provide your reasoning, the bounding box of the lateral meniscus, and the location of any tear. Tear location is encoded as one of the following integers: \n0 = LMA (Lateral Meniscus Anterior), \n1 = LMB (Lateral Meniscus Body), \n2 = LMP (Lateral Meniscus Posterior), 3 = No tear detected. The output format should be a JSON object with the bounding box coordinates and the cyst presence: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"lateral_meniscal_cyst\": {\"presence\": 0 | 1, \"tear_location\": (0-3)}}.",
+  "56": "In this DESS MRI, detect medial meniscal hypertrophy, if there is any. Provide your reasoning, the bounding box of the medial meniscus region, and whether it's present. The output format should be a JSON object with the bounding box coordinates and the hypertrophy presence: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"medial_meniscal_hypertrophy\": {\"presence\": (0 | 1)}}",
+  "57": "In this DESS MRI, detect lateral meniscal hypertrophy, if there is any. Provide your reasoning, the bounding box of the lateral meniscus region, and whether it's present. The output format should be a JSON object with the bounding box coordinates and the hypertrophy presence: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"lateral_meniscal_hypertrophy\": {\"presence\": (0 | 1)}}",
+  "58": "In this DESS MRI, detect a meniscal root tear at the MMRT, if there is any. Provide your reasoning and the bounding box of the medial meniscus region. The output format should be a JSON object with the bounding box coordinates and the root tear information: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"medial_meniscal_root_tear\": {\"presence\": (0 | 1)}}",
+  "59": "In this DESS MRI, detect a meniscal root tear at the LMRT, if there is any. Provide your reasoning and the bounding box of the lateral meniscus region. The output format should be a JSON object with the bounding box coordinates and the root tear information: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"lateral_meniscal_root_tear\": {\"presence\": (0 | 1)}}",
+  "60": "In this DESS MRI, assess the ACL. Provide your reasoning, the bounding box of the ACL region and indicate for the following: whether there is a complete tear and whether there is an associated BML or cyst at the site of insertion or origin, and whether there has been ACL repair. Provide your reasoning, the bounding box of the ACL region, and the assessment. The output format should be a JSON object with the bounding box coordinates and the assessment: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"ACL\": {\"Complete Tear\": (0 | 1), \"Associated with BML/cyst at site of insertion or origin\": (0 | 1), \"ACL Repair\": (0 | 1)}}",
+  "61": "In this DESS MRI, assess the PCL. Provide your reasoning, the bounding box of the PCL region and indicate for the following: whether there is a complete tear and whether there is an associated BML or cyst at the site of insertion or origin. Provide your reasoning, the bounding box of the PCL region, and the assessment. The output format should be a JSON object with the bounding box coordinates and the assessment: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"PCL\": {\"Complete Tear\": (0 | 1), \"Associated with BML/cyst at site of insertion or origin\": (0 | 1)}}",
+  "62": "In this DESS MRI, assess the patellar tendon. Provide your reasoning, the bounding box of the patellar tendon region, and whether there is a signal abnormality. Provide your reasoning, the bounding box of the patellar tendon region, and the assessment. The output format should be a JSON object with the bounding box coordinates and the assessment: {\"bounding_box\": [start_slice, end_slice, x_min, x_max, y_min, y_max], \"Patellar_Tendon\": {\"Signal Abnormality\": (0 | 1)}}",
+  "63": "In this DESS MRI, detect pes anserine bursitis, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the pes anserine bursitis presence: {\"pes_anserine_bursitis_presence\": {\"presence\": (0 | 1)}}",
+  "64": "In this DESS MRI, detect iliotibial band signal, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the iliotibial band signal presence: {\"iliotibial_band_signal_presence\": {\"presence\": (0 | 1)}}",
+  "65": "In this DESS MRI, detect a popliteal cyst, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the popliteal cyst presence: {\"popliteal_cyst_presence\": {\"presence\": (0 | 1)}}",
+  "66": "In this DESS MRI, detect infrapatellar bursitis, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the infrapatellar bursitis presence: {\"infrapatellar_bursitis_presence\": {\"presence\": (0 | 1)}}",
+  "67": "In this DESS MRI, detect prepatellar bursitis, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the prepatellar bursitis presence: {\"prepatellar_bursitis_presence\": {\"presence\": (0 | 1)}}",
+  "68": "In this DESS MRI, detect a ganglion cyst, if there is any. Provide your reasoning and whether it is present. The output format should be a JSON object with the ganglion cyst presence: {\"ganglion_cyst_presence\": {\"presence\": (0 | 1)}}",
+  "69": "In this DESS MRI, detect loose bodies, if there are any. Provide your reasoning and whether it is present. The output format should be a JSON object with the loose bodies presence: {\"loose_bodies_presence\": {\"presence\": (0 | 1)}}"
+}