PULPo: Probabilistic Unsupervised Laplacian Pyramid Registration

A PyTorch Lightning implementation of PULPo for probabilistic deformable medical image registration with uncertainty quantification using a hierarchical variational autoencoder (VAE) architecture. This repository contains the code for our publication at the MICCAI conference in 2024 (https://conferences.miccai.org/2024).

Paper: PULPo: Probabilistic Unsupervised Laplacian Pyramid Registration
Preprint: arXiv:2407.10567
Research Group: [ML in Medical Image Analysis] (https://www.mlmia-unitue.de/)

Citation

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

@Inproceedings{10.1007/978-3-031-72069-7_67,
  title="PULPo: Probabilistic Unsupervised Laplacian Pyramid Registration",
  author="Siegert, Leonard
and Fischer, Paul
and Heinrich, Mattias P.
and Baumgartner, Christian F.",
  booktitle="Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024",
  year="2024",
  pages="717--727",
  isbn="978-3-031-72069-7"
}