multiGradICON: A Foundation Model for Multimodal Medical Image Registration

TL;DR

multiGradICON introduces a deep learning model capable of accurate, fast, and universal multimodal medical image registration, advancing beyond previous monomodal-focused models and demonstrating improved generalization and accuracy.

Contribution

It develops the first deep learning model for universal multimodal medical image registration, enhancing accuracy and generalization over existing monomodal models.

Findings

Training with multimodal data improves generalization.

Loss function randomization boosts registration accuracy.

Model performs well on both monomodal and multimodal registration tasks.

Abstract

Modern medical image registration approaches predict deformations using deep networks. These approaches achieve state-of-the-art (SOTA) registration accuracy and are generally fast. However, deep learning (DL) approaches are, in contrast to conventional non-deep-learning-based approaches, anatomy-specific. Recently, a universal deep registration approach, uniGradICON, has been proposed. However, uniGradICON focuses on monomodal image registration. In this work, we therefore develop multiGradICON as a first step towards universal *multimodal* medical image registration. Specifically, we show that 1) we can train a DL registration model that is suitable for monomodal *and* multimodal registration; 2) loss function randomization can increase multimodal registration accuracy; and 3) training a model with multimodal data helps multimodal generalization. Our code and the multiGradICON model are available at https://github.com/uncbiag/uniGradICON.