Stationary Velocity Fields on Matrix Groups for Deformable Image Registration

TL;DR

This paper extends the stationary velocity field approach to matrix groups like SE(3), enabling better handling of large deformations in deformable image registration by leveraging low-frequency transformations.

Contribution

It introduces a matrix group extension of SVF, provides conditions for flow existence, and proposes an efficient numerical integration method for large deformation registration.

Findings

Successfully applied to 3D MRI brain images

Improved registration of large deformations

Validated with numerical experiments

Abstract

The stationary velocity field (SVF) approach allows to build parametrizations of invertible deformation fields, which is often a desirable property in image registration. Its expressiveness is particularly attractive when used as a block following a machine learning-inspired network. However, it can struggle with large deformations. We extend the SVF approach to matrix groups, in particular $\SE(3)$. This moves Euclidean transformations into the low-frequency part, towards which network architectures are often naturally biased, so that larger motions can be recovered more easily. This requires an extension of the flow equation, for which we provide sufficient conditions for existence. We further prove a decomposition condition that allows us to apply a scaling-and-squaring approach for efficient numerical integration of the flow equation. We numerically validate the approach on inter-patient registration of 3D MRI images of the human brain.