Explicit-Shape-Priors

This repo is the official implementation for: Learning with Explicit Shape Priors for Medical Image Segmentation

Dataset Link

BraTS 2020: Multimodal Brain Tumor Segmentation Challenge 2020

VerSe'19: Large Scale Vertebrae Segmentation Challenge

Automated Cardiac Diagnosis Challenge (ACDC)

Preprocess

We follow the z-score normalization strategy in nnUNet to preprocess the BraTS 2020, VerSe'19 and ACDC dataset.

Requirements

  • python 3.7
  • pytorch 1.8.0
  • torchvision 0.9.0
  • simpleitk 2.0.2
  • monai 0.9.0

Training

If you want to train the model from scratch, run the training script as following.
python BraTS_train.py
python VerSe_train.py
python ACDC_train.py

Testing

If you want to test the model, run the testing script as following.
python BraTS_test.py
python VerSe_test.py
python ACDC_test.py

Citation

If you use our code or models in your work or find it is helpful, please cite the corresponding paper:

@article{you2023learning,
  title={Learning with Explicit Shape Priors for Medical Image Segmentation},
  author={You, Xin and He, Junjun and Yang, Jie and Gu, Yun},
  journal={arXiv preprint arXiv:2303.17967},
  year={2023}
}