fireants 1.0.0

FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Registration

:fire: FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Registration

The FireANTs library is a lightweight registration package for Riemannian diffeomorphic registration on GPUs.

Installation

To use the FireANTs package, you can either

• Use Docker (see docker/README.md, highly recommended)
• Clone the repository and install the package locally
• Install the package directly from PyPI.

If you do not want to use Docker, we recommend using a fresh Anaconda/Miniconda environment to install the package.

conda create -n fireants python=3.9

To install FireANTs locally:

git clone https://github.com/rohitrango/fireants
cd fireants
pip install .
pip install ./fused_ops

By default, doing pip install . will NOT include the fused CUDA operations. I highly recommend installing the fused CUDA operations using pip install ./fused_ops.

Or install from PyPI:

pip install fireants
pip install fireants_fused_ops

Tutorial

To check out some of the tutorials, check out the tutorials/ directory for usage. Alternatively, to reproduce the results in the paper checkout the fireants/scripts/ directory.

Documentation

You can also check out the Documentation. Feel free to reach out to me for improvements in the documentation.

Datasets

In the paper, we use the datasets as following:

• Klein's evaluation of 14 non-linear registration algorithms: here
• EMPIRE10 lung registration challenge: here
• Expansion Microscopy dataset: here

Contributing

Feel free to add issues or pull requests to the repository. We welcome contributions to the package.

Tests

To run the tests, run the following command:

bash run_tests.sh

Feel free to report any issues with the tests to the issues page.

License

Please refer to the LICENSE file for the license details, especially pertaining to redistribution of code and derivative works.

Citation

If you use FireANTs in your research, please cite the following paper:

@article{jena2024fireants,
  title={FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Registration},
  author={Jena, Rohit and Chaudhari, Pratik and Gee, James C},
  journal={arXiv preprint arXiv:2404.01249},
  year={2024}
}

If you use FireANTs-as-a-layer (Deep Implicit Optimization, code), cite the following paper:

@article{jena2025deep,
  title={Deep implicit optimization enables robust learnable features for deformable image registration},
  author={Jena, Rohit and Chaudhari, Pratik and Gee, James C},
  journal={Medical Image Analysis},
  volume={103},
  pages={103577},
  year={2025},
  publisher={Elsevier}
}