Home

Awesome

Semi-supervised medical image segmentation using adversarial consistency learning and dynamic convolution network

#ASE-Net

Usage

  1. Clone the repo.;
git clone https://github.com/SUST-reynole/ASE-Net.git
  1. Put the data in './data';

  2. Train the model;

cd ASE-Net
# e.g., for 20% labels on LA
python3 ./code/train_ASE-Net_3D.py --dataset_name LA  --labelnum 16 --gpu 0 --temperature 0.1
  1. Test the model;
cd ASE-Net
# e.g., for 20% labels on LA
python ./code/test_LA_our.py --dataset_name LA  --labelnum 16 --gpu 0

Acknowledgements:

Our code is origin from UAMT, SASSNet, DTC, URPC, SSL4MIS and MC-Net. Thanks for these authors for their valuable works and hope our model can promote the relevant research as well.

Questions