SANet

Exploring Separable Attention for Multi-Contrast MR Image Super-Resolution

Dependencies

• numpy==1.18.5
• scikit_image==0.16.2
• torchvision==0.8.1
• torch==1.7.0
• runstats==1.8.0
• pytorch_lightning==1.0.6
• h5py==2.10.0
• PyYAML==5.4

Train

cd experimental/SANet/
sbatch job.sh

Change other arguments that you can train your own model.

Parameter settings

The detailed parameter settings can be find in our arXiv paper: https://arxiv.org/pdf/2109.01664.pdf

Citation

If you find SANet useful for your research, please consider citing the following papers:

@inproceedings{feng2021MINet,
  title={Multi-Contrast MRI Super-Resolution via a Multi-Stage Integration Network},
  author={Feng, Chun-Mei and Fu, Huazhu and Yuan, Shuhao and Xu, Yong},
  booktitle={International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)},
  year={2021}
}
@article{feng2021exploring,
  title={Exploring Separable Attention for Multi-Contrast MR Image Super-Resolution},
  author={Feng, Chun-Mei and Yan, Yunlu and Liu, Chengliang and Fu, Huazhu and Xu, Yong and Shao, Ling},
  journal={arXiv preprint arXiv:2109.01664},
  year={2021}
}