MAD: Modality Agnostic Distance Measure for Image Registration

TL;DR

MAD is a novel deep image distance measure that uses random convolutions to enable accurate, modality-agnostic registration of multi-modal images, trained on mono-modal data and effective across various modalities.

Contribution

Introduces MAD, a modality-agnostic deep distance measure utilizing random convolutions, allowing training on mono-modal data and successful application to multi-modal image registration.

Findings

MAD achieves successful affine registration of multi-modal images.

MAD has a larger capture range than traditional measures.

MAD is robust to large appearance changes across modalities.

Abstract

Multi-modal image registration is a crucial pre-processing step in many medical applications. However, it is a challenging task due to the complex intensity relationships between different imaging modalities, which can result in large discrepancy in image appearance. The success of multi-modal image registration, whether it is conventional or learning based, is predicated upon the choice of an appropriate distance (or similarity) measure. Particularly, deep learning registration algorithms lack in accuracy or even fail completely when attempting to register data from an "unseen" modality. In this work, we present Modality Agnostic Distance (MAD), a deep image distance}] measure that utilises random convolutions to learn the inherent geometry of the images while being robust to large appearance changes. Random convolutions are geometry-preserving modules which we use to simulate an infinite number of synthetic modalities alleviating the need for aligned paired data during training. We can therefore train MAD on a mono-modal dataset and successfully apply it to a multi-modal dataset. We demonstrate that not only can MAD affinely register multi-modal images successfully, but it has also a larger capture range than traditional measures such as Mutual Information and Normalised Gradient Fields.