HiLab-git / WSL4MIS

Scribbles or Points-based weakly-supervised learning for medical image segmentation, a strong baseline, and tutorial for research and application.
MIT License
198 stars 34 forks source link
medical-image-segmentation weakly-supervised-learning weakly-supervised-segmentation

Weakly-supervised learning for medical image segmentation (WSL4MIS).

Dataset

Follow official guidance to install Pytorch.

Usage

  1. Clone this project.

    git clone https://github.com/HiLab-git/WSL4MIS
    cd WSL4MIS
  2. Data pre-processing os used or the processed data.

    cd code
    python dataloaders/acdc_data_processing.py
  3. Train the model

    cd code
    bash train_wss.sh # train model with scribble or dense annotations.
    bash train_ssl.sh  # train model with mix-supervision (mask annotations and without annotation).
  4. Test the model

    python test_2D_fully.py --sup_type scribble/label --exp ACDC/the trained model fold --model unet
    python test_2D_fully_sps.py --sup_type scribble --exp ACDC/the trained model fold --model unet_cct
  5. Training curves on the fold1: Note: pCE means partially cross-entropy, TV means total variation, label denotes supervised by mask, scribble represents just supervised by scribbles.

Implemented methods

Acknowledgement