TransMorph with Deformable Cross-attention (DCA)

[PyTorch] Deformable Cross-Attention Transformer for Medical Image Registration

This is a PyTorch implementation of my paper:
Accepted to MLMI 2023!
Chen, Junyu, et al. "Deformable Cross-Attention Transformer for Medical Image Registration." In Machine Learning in Medical Imaging: 14th International Workshop, MLMI 2023.

We provide pretrained weights, as well as training and inference scripts on two datasets, IXI and OASIS.
OASIS Scripts: Code
IXI Scripts: Code

Deformable cross-attention illustration:

TransMorph-DCA Architecture:

Example Results:

Citation:

If you find this code is useful in your research, please consider to cite:

@article{chen2023deformable,
title={Deformable Cross-Attention Transformer for Medical Image Registration},
author={Chen, Junyu and Liu, Yihao and He, Yufan and Du, Yong},
journal={arXiv preprint arXiv:2303.06179},
year={2023}
}