/kidney-sonography

kidney function classification and prediction through ultrasound-based kidney imaging: from deep learning to mass screening of chronic kidney disease

Primary LanguageJupyter Notebook

kidney-sonography

Introduction

kidney function classification and prediction through ultrasound-based kidney imaging: from deep learning to mass screening of chronic kidney disease

Prediction module

Ensembles 10 trained models to predict estimated glomerular filtration rate (eGFR).

Tutorial

  1. Run get_models.sh to retrieve 10 trained model weights. (Warning! The file size is about 1.7 GB).
  2. Execute ensemble_predict.py
  3. Input file path to a cropped kidney sonography. Sample images are provided in "/sample_images".

Use --help to see usage of ensemble_predict.py:

usage: ensemble_predict.py [-h] [-g]

optional arguments:
-h, --help    show help message and exit
-g, --gpu_id  assign GPU ID, default 0