supergengchen / nCoVSegNet

The code for 'COVID-19 Lung Infection Segmentation with A Novel Two-Stage Cross-Domain Transfer Learning Framework'

nCoVSegNet

COVID-19 Lung Infection Segmentation with A Novel Two-Stage Cross-Domain Transfer Learning Framework

Requirements

• Python3
• Pytorch version >= 1.2.0.
• Some basic python packages, such as Numpy, Pandas, SimpleITK.

Data Preparation

• Please put CT images and segmentation masks in the following directory: ./dataset/, and organize the data as follows:

     ├── dataset
        ├── train
           ├── image
               ├── 1.jpg, 2.jpg, xxxx
           ├── mask
               ├── 1.png, 2.png, xxxx
        ├── test
           ├── image
               ├── case01
                   ├── 1.jpg, 2.jpg, xxxx
               ├── xxxx
           ├── mask
               ├── case01
                   ├── 1.png, 2.png, xxxx
               ├── xxxx
  

Training & Testing

• Train the nCoVSegNet:

  python train.py

• Test the nCoVSegNet:

  python test.py

  The results will be saved to ./Results.

• Evaluate the segmentation maps:

  You can evaluate the segmentation maps using the tool in ./utils/evaluation.py.

Citation

Please cite the following paper if you use this repository in your reseach.

@article{liu2021covid19,
title={COVID-19 Lung Infection Segmentation with A Novel Two-Stage Cross-Domain Transfer Learning Framework},
author={Jiannan Liu, Bo Dong, Shuai Wang, Hui Cui, Dengping Fan, Jiquan Ma, Geng Chen},
booktitle={Medical Image Analysis},
year={2021}
}

Acknowledgement

A collection of COVID-19 imaging-based AI research papers and datasets: https://github.com/HzFu/COVID19_imaging_AI_paper_list

License

Our code is released under MIT License (see LICENSE file for details).