metadata

license: gpl-3.0

Breast Estrogen Receptor v1 Model Card

This model card describes the model associated with the manuscript "Uncertainty-informed deep learning models enable high-confidence predictions for digital histopathology", by Dolezal et al, available here.

Model Details

Developed by: James Dolezal
Model type: Deep convolutional neural network image classifier
Language(s): English
License: GPL-3.0
Model Description: This is a model that can predict, from H&E-stained pathologic images of breast cancer, whether a tumor is likely to be estrogen receptor (ER) negative or positive. It is an Xception model with two dropout-enabled hidden layers.
Image processing: This model expects images of H&E-stained pathology slides at 299 x 299 px and 302 x 302 μm resolution. Images should be stain-normalized using a modified Reinhard normalizer ("Reinhard-Fast") available here. The stain normalizer should be fit using the target_means and target_stds listed in the model params.json file. Images should be should be standardized with tf.image.per_image_standardization().
Resources for more information: GitHub Repository

Uses

Examples

For direct use, the model can be loaded using Tensorflow/Keras:

import tensorflow as tf
model = tf.keras.models.load_model('/path/')

or loaded with Slideflow version 1.1+ with the following syntax:

import slideflow as sf
model = sf.model.load('/path/')

The stain normalizer can be loaded and fit using Slideflow:

normalizer = sf.util.get_model_normalizer('/path/')

The stain normalizer has a native Tensorflow transform and can be directly applied to a tf.data.Dataset:

# Map the stain normalizer transformation
# to a tf.data.Dataset
dataset = dataset.map(normalizer.tf_to_tf)

Alternatively, the model can be used to generate predictions for whole-slide images processed through Slideflow in an end-to-end Project. To use the model to generate predictions on data processed with Slideflow, simply pass the model to the Project.predict() function:

import slideflow
P = sf.Project('/path/to/slideflow/project')
P.predict('/model/path')

Direct Use

This model is intended for research purposes only. Possible research areas and tasks include

Applications in educational settings.
Research on pathology classification models for breast cancer.

Excluded uses are described below.

Misuse and Out-of-Scope Use

This model should not be used in a clinical setting to generate predictions that will be used to inform patients, physicians, or any other health care members directly involved in their health care outside the context of an approved research protocol. Using the model in a clinical setting outside the context of an approved research protocol is a misuse of this model. This includes, but is not limited to:

Generating predictions of images from a patient's tumor and sharing those predictions with the patient
Generating predictions of images from a patient's tumor and sharing those predictions with the patient's physician, or other members of the patient's healthcare team
Influencing a patient's health care treatment in any way based on output from this model

Limitations

The model has not been validated to discriminate estrogen receptor status in a manner which controls for possible underlying biological bias, such tumor grade or histological subtype.

Bias

This model was trained on The Cancer Genome Atlas (TCGA), which contains patient data from communities and cultures which may not reflect the general population. This datasets is comprised of images from multiple institutions, which may introduce a potential source of bias from site-specific batch effects (Howard, 2021).

Training

Training Data The following dataset was used to train the model:

The Cancer Genome Atlas (TCGA), BRCA cohort (see next section)

This model was trained on a total of 1,048 slides, with 228 ER-negative tumor and 820 ER-positive tumors.

Training Procedure Each whole-slide image was sectioned into smaller images in a grid-wise fashion in order to extract tiles from whole-slide images at 302 x 302 μm. Image tiles were extracted at the nearest downsample layer, and resized to 299 x 299 px using Libvips. During training,

Images are stain-normalized with a modified Reinhard normalizer ("Reinhard-Fast"), which excludes the brightness standardization step, available here
Images are randomly flipped and rotated (90, 180, 270)
Images have a 50% chance of being JPEG compressed with quality level between 50-100%
Images have a 10% chance of random Gaussian blur, with sigma between 0.5-2.0
Images are standardized with tf.image.per_image_standardization()
Images are classified through an Xception block, followed by two hidden layers with dropout (p=0.1) enabled during training
The loss is cross-entropy, with ER-negative=0 and ER-positive=1
Training is completed after 1 epoch

Additional training information:

Hardware: 1 x A100 GPUs
Optimizer: Adam
Batch: 128
Learning rate: 0.0001, with a decay of 0.98 every 512 steps
Hidden layers: 2 hidden layers of width 1024, with dropout p=0.1

Evaluation Results

External evaluation results are currently under peer review and will be posted once publicly available.