Unsupervised Similarity Learning for Image Registration With Energy-Based Models

MCML Authors

Mohammad Farid Azampour

→ Group Nassir Navab
 Computer Aided Medical Procedures & Augmented Reality

Nassir Navab

Prof. Dr.

Principal Investigator

Computer Aided Medical Procedures & Augmented Reality

Julia Schnabel

Prof. Dr.

Principal Investigator

Computational Imaging and AI in Medicine

Abstract

We present a new model for deformable image registration, which learns in an unsupervised way a data-specific similarity metric. The proposed method consists of two neural networks, one that maps pairs of input images to transformations which align them, and one that provides the similarity metric whose maximisation guides the image alignment. We parametrise the similarity metric as an energy-based model, which is simple to train and allows us to improve the accuracy of image registration compared to other models with learnt similarity metrics by taking advantage of a more general mathematical formulation, as well as larger datasets. We also achieve substantial improvement in the accuracy of inter-patient image registration on MRI scans from the OASIS dataset compared to models that rely on traditional functions.

inproceedings

---

WBIR @MICCAI 2024

11th International Workshop on Biomedical Image Registration at the 27th International Conference on Medical Image Computing and Computer Assisted Intervention. Marrakesh, Morocco, Oct 06-10, 2024.

---

Authors

D. Grzech • L. Le Folgoc • M. F. Azampour • A. Vlontzos • B. Glocker • N. Navab • J. A. Schnabel • B. Kainz

---

Links

DOI

---

Research Area

 C1 | Medicine