Skip to content

Latest commit

 

History

History
executable file
·
25 lines (17 loc) · 1.26 KB

File metadata and controls

executable file
·
25 lines (17 loc) · 1.26 KB

Multi-modal unsupervised brain image registration using edge maps

This repository contains the code described in the paper Multi-modal unsupervised brain image registration using edge maps.

The current work extends the one of Qui et al. Learning Diffeomorphic and Modality-invariant Registrationusing B-splines. For more information on the original models, the file structure, configurations, training your own model, hyper-parameter tuning and inference, please consider visiting their repository: https://github.com/qiuhuaqi/midir.

Please consider citing the paper if you use the code in this repository.

The image presents an overview of the method.

Installation

  1. Clone this repository
  2. In a fresh Python 3.8.x virtual environment, install dependencies via:
    pip install -r <path_to_cloned_repository>/requirements.txt
    

Comments

We tested the code on a Quadro RTX 8000 GPU using CUDA 11.3 and CuDNN 8.3.

Contact Us

If you have any question or need any help running the code, feel free to open an issue or email us at: vasiliki.sideri-lampretsa@tum.de