Toward Universal Medical Image Registration via Sharpness-Aware Meta-Continual Learning

TL;DR

This paper introduces a sharpness-aware meta-continual learning approach for universal 3D medical image registration, effectively handling distribution shifts and enabling registration across diverse tasks in sequential learning scenarios.

Contribution

It proposes a novel sharpness-aware meta-continual learning method that improves generalization and mitigates catastrophic forgetting in universal medical image registration.

Findings

SAMCL outperforms vanilla sequential training.

Method achieves comparable results to multi-task training.

Validated on four diverse medical imaging datasets.

Abstract

Current deep learning approaches in medical image registration usually face the challenges of distribution shift and data collection, hindering real-world deployment. In contrast, universal medical image registration aims to perform registration on a wide range of clinically relevant tasks simultaneously, thus having tremendous potential for clinical applications. In this paper, we present the first attempt to achieve the goal of universal 3D medical image registration in sequential learning scenarios by proposing a continual learning method. Specifically, we utilize meta-learning with experience replay to mitigating the problem of catastrophic forgetting. To promote the generalizability of meta-continual learning, we further propose sharpness-aware meta-continual learning (SAMCL). We validate the effectiveness of our method on four datasets in a continual learning setup, including brain MR, abdomen CT, lung CT, and abdomen MR-CT image pairs. Results have shown the potential of SAMCL in realizing universal image registration, which performs better than or on par with vanilla sequential or centralized multi-task training strategies.The source code will be available from https://github.com/xzluo97/Continual-Reg.