Skip to content

Latest commit

 

History

History
112 lines (76 loc) · 4.47 KB

README.md

File metadata and controls

112 lines (76 loc) · 4.47 KB

CoCoLIT (AAAI'26)

PyTorch Paper PDF Hugging Face Model

Open in Google Colab

CoCoLIT: ControlNet-Conditioned Latent Image Translation for MRI to Amyloid PET Synthesis
Alec Sargood*, Lemuel Puglisi*, James Cole, Neil Oxtoby, Daniele Ravì†, Daniel C. Alexander†
* Joint first authors, † Joint senior authors

News

  • 2026-03 Try CoCoLIT on your own data using our notebook on Google Colab 🔥
  • 2025-10: Our paper CoCoLIT has been accepted at AAAI 2026 (17% acceptance rate).

Installation

This repository requires Python >= 3.10 & <= 3.13. To install the latest version, run:

pip install cocolit

Usage

After installing the package, you can convert a T1-weighted MRI to a Florbetapir SUVR map by running:

mri2pet --i /path/to/t1.nii.gz --o /path/to/output.nii.gz

If your MRI has not been preprocessed yet, you can run the full preprocessing pipeline directly from the CLI (skull-stripping, affine registration to MNI space, and WhiteStripe normalization):

mri2pet --i /path/to/t1.nii.gz --o /path/to/output.nii.gz \
  --preprocess

Optionally, save intermediate preprocessing files with:

mri2pet --i /path/to/t1.nii.gz --o /path/to/output.nii.gz \
  --preprocess \
  --preprocess_out /path/to/preproc_outputs

To replicate the results presented in the paper, include the --m 64 flag.

Training & Reproducibility

Note

If you wish to compare your MRI-to-PET model against CoCoLIT but prefer not to retrain our models due to limited compute or time, please email us and we will provide the exact training splits used in this work to enable benchmarking with our pretrained models.

To reproduce the experiments reported in the paper, please follow the reproducibility guide.

Disclaimer

This software is not intended for clinical use. The code is not available for commercial applications. For commercial inquiries, please contact the corresponding authors.

License: Attribution-NonCommercial 4.0 International

Citing

Arxiv Preprint:

@article{sargood2025cocolit,
  title={CoCoLIT: ControlNet-Conditioned Latent Image Translation for MRI to Amyloid PET Synthesis},
  author={Sargood, Alec and Puglisi, Lemuel and Cole, James H and Oxtoby, Neil P and Rav{\`\i}, Daniele and Alexander, Daniel C},
  journal={arXiv preprint arXiv:2508.01292},
  year={2025}
}

Contacts

For any inquiries, please contact: