# Metric Learning for Image Registration

## TL;DR

This paper introduces a novel deep learning-based regularizer for image registration that allows adaptive control over transformation regularity, improving structural preservation and enabling diffeomorphic transformations.

## Contribution

It proposes embedding a deep learning model within an optimization-based registration framework to learn a spatially-adaptive regularizer, a significant departure from existing methods.

## Key findings

- Enables control over the regularity of transformations.
- Allows preservation of structural properties like diffeomorphism.
- Integrates deep learning with traditional optimization for registration.

## Abstract

Image registration is a key technique in medical image analysis to estimate deformations between image pairs. A good deformation model is important for high-quality estimates. However, most existing approaches use ad-hoc deformation models chosen for mathematical convenience rather than to capture observed data variation. Recent deep learning approaches learn deformation models directly from data. However, they provide limited control over the spatial regularity of transformations. Instead of learning the entire registration approach, we learn a spatially-adaptive regularizer within a registration model. This allows controlling the desired level of regularity and preserving structural properties of a registration model. For example, diffeomorphic transformations can be attained. Our approach is a radical departure from existing deep learning approaches to image registration by embedding a deep learning model in an optimization-based registration algorithm to parameterize and data-adapt the registration model itself.