This project is developed for our MICCAI 2024 paper: ModelMix: A New Model-Mixup Strategy to Minimize Vicinal Risk across Tasks for Few-scribble based Cardiac Segmentation. For more information about ModelMix, please read the following paper:

@article{zhang2024modelmix,
  title={ModelMix: A New Model-Mixup Strategy to Minimize Vicinal Risk across Tasks for Few-scribble based Cardiac Segmentation},
  author={Zhang, Ke and Patel, Vishal M},
  journal={arXiv preprint arXiv:2406.13237},
  year={2024}
}

Please also cite this paper if you are using ModelMix for your research.

1. The MSCMR dataset with mask annotations can be downloaded from MSCMRseg.
2. The scribble annotations of MSCMRseg have been released in MSCMR_scribbles.
3. The scribble-annotated MSCMR dataset used for training could be directly downloaded from MSCMR_dataset.
4. The ACDC dataset with mask annotations can be downloaded from ACDC and the scribble annotations could be downloaded from ACDC scribbles. You can also directly download the [scribble-annotated ACDC dataset].(https://github.com/BWGZK/ModelMix/tree/main/ACDC_dataset)
5. The MyoPS dataset can be downloaded in MyoPS dataset
6. The scribbles of MyoPS dataset have been released in MyoPS scribbles
7. Please organize the dataset as the following structure:

XXX_dataset/
  -- TestSet/
      --images/
      --labels/
  -- train/
      --images/
      --labels/
  -- val/
      --images/
      --labels/

1. Set the "dataset" parameter in main.py, line 76, to the name of dataset, i.e., "MSCMR_dataset".
2. Set the "output_dir" in main.py, line 79, as the path to save the checkpoints.
3. Download the dataset, for example, the MSCMR_dataset. Then, Set the dataset path in /data/mscmr.py, line 110, to your data path where the dataset is located in.
4. Check your GPU devices and modify the "GPU_ids" parameter in main.py, line 83 and "CUDA_VISIBLE_DEVICES" in run.sh.
5. Start to train by:.

python main.py

pip install -r requirement.txt

If you have any problems, please feel free to contact us. Thanks for your attention.