Official repository for "OTMorph: Unsupervised Multi-domain Abdominal Medical Image Registration Using Neural Optimal Transport", published in IEEE Transactions on Medical Imaging.

[IEEE TMI paper]

To test the proposed OTMorph, we may need Conda environment. You can download Anaconda from this link. Then, please run:

conda env create –f OTMorph.yml

To train our model, run this command:

sh ./scripts/train.sh

To test the trained our model, run:

sh ./scripts/test.sh

When you run the code, "./results" folder is generated. And the registration results are saved in this directory.

@article{kim2024otmorph,
  title={OTMorph: Unsupervised Multi-domain Abdominal Medical Image Registration Using Neural Optimal Transport},
  author={Kim, Boah and Zhuang, Yan and Mathai, Tejas Sudharshan and Summers, Ronald M},
  journal={IEEE Transactions on Medical Imaging},
  year={2024},
  publisher={IEEE}
}