This is a fork of a Pytorch meant to customize it for my own purposes. Please see other versions of the project to customize it for your needs.

This is a Python3 (Pytorch) reimplementation of CheXNet. The model takes a chest X-ray image as input and outputs the probability of certain thoracic pathologies along with a likelihood map of pathologies.

The original dataset for the original project:

The ChestX-ray14 dataset comprises 112,120 frontal-view chest X-ray images of 30,805 unique patients with 14 disease labels. To evaluate the model, we randomly split the dataset into training (70%), validation (10%) and test (20%) sets, following the work in paper. Partitioned image names and corresponding labels are placed under the directory labels.

• Python 3.4+
• PyTorch and its dependencies

0. I would clone a repository without my modifications...but if you want:

1. Clone this repository.

2. Download images of ChestX-ray14 from this released page and decompress them to the directory images.

3. Specify one or multiple GPUs and run

   python model.py

This work is based on earllier work collaboratively conducted by Xinyu Weng, Nan Zhuang, Jingjing Tian and Yingcheng Liu. They were students/interns of Machine Intelligence Lab, Institute of Computer Science & Technology, Peking University, directed by Prof. Yadong Mu (http://www.muyadong.com). They in tern based their work off the original CheXnet project out of Stanford.