SmileCode

The publicly available code for ModeT and the other medical image registration codes released by the Smile Lab.

ModeT: Learning Deformable Image Registration via Motion Decomposition Transformer (MICCAI2023)

By Haiqiao Wang, Dong Ni, Yi Wang.

(03/12/2024 News) We have uploaded ModeT integrated with the GPU operator of ModeTv2 (ModeT-cu), which greatly reduces execution time and memory requirements. For further instructions and reference, please refer to [ModeTv2].

Paper link: [MICCAI]

Environment

Code has been tested with Python 3.9 and PyTorch 1.11.

Dataset

LPBA [link] Mindboggle [link]

Instruction

For convenience, we are sharing the preprocessed LPBA dataset used in our experiments. Once uncompressed, simply modify the "LPBA_path" in train.py to the path name of the extracted data. Next, you can execute train.py to train the network, and after training, you can run infer.py to test the network performance.

(Update) We encourage you to try the ModeTv2 code, as it enhances registration accuracy while significantly reducing both runtime and memory usage.

Citation

If you use the code in your research, please cite:

@InProceedings{10.1007/978-3-031-43999-5_70,
author="Wang, Haiqiao and Ni, Dongand Wang, Yi",
title="ModeT: Learning Deformable Image Registration via Motion Decomposition Transformer",
booktitle="Medical Image Computing and Computer Assisted Intervention -- MICCAI 2023",
year="2023",
pages="740--749",
}

The overall framework and some network components of the code are heavily based on TransMorph and VoxelMorph. We are very grateful for their contributions. The file makePklDataset.py shows how to make a pkl dataset from the original LPBA dataset. If you have any other questions about the .pkl format, please refer to the github page of [TransMorph_on_IXI].

Other Works

ModeTv2: GPU-accelerated Motion Decomposition Transformer for Pairwise Optimization in Medical Image Registration

By Haiqiao Wang, Zhuoyuan Wang, Dong Ni, Yi Wang.

Paper link: [arxiv], Code link: [code]

Recursive Deformable Pyramid Network for Unsupervised Medical Image Registration (TMI2024)

By Haiqiao Wang, Dong Ni, Yi Wang.

Paper link: [TMI], Code link: [code]

Unofficial Pytorch implementations (Baseline Methods)

• Recursive Cascaded Networks for Unsupervised Medical Image Registration (RCN)

  links: [original code] [paper] [code]

• Recursive Decomposition Network for Deformable Image Registration (RDN)

  links: [paper] [code]

• Joint Progressive and Coarse-to-Fine Registration of Brain MRI via Deformation Field Integration and Non-Rigid Feature Fusion (PCnet)

  links: [original code] [paper] [code]

• Dual-stream pyramid registration network (PR++)

  links: [paper] [code]

• Coordinate Translator for Learning Deformable Medical Image Registration (Im2Grid)

  links: [paper] [code]

How can other datasets be used in this code?

This is a common question, and please refer to the github page of ChangeDataset.md for more information.