Skip to content

Implementation of paper ‘Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models” in Tensorflow.

Notifications You must be signed in to change notification settings

yueliukth/decoupling_breast_cancer_risk

Repository files navigation

Repository for MICCAI submission

"Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models”.

Alt text

Implemented in TensorFlow.

Main scripts

preprocessing.py		Mammogram preprocessing from dicom to 16-bit png
tfbuilder.py 			Store input data in TFRecord format
main_run.py 			Execute the main scripts 

Scripts to support the main execution script ‘main_run.py’

exp.py 				Define model hyperparameters 
train_network.py		Define training process
eval_network.py		        Define testing process
dataset_functions.py	        Define dataset helping functions
train_functions.py		Define training helping functions
model_functions.py		Define ResNet model

About

Implementation of paper ‘Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models” in Tensorflow.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages