Automated Learning for Deformable Medical Image Registration by Jointly Optimizing Network Architectures and Objective Functions

TL;DR

This paper introduces AutoReg, an automated learning framework that jointly optimizes network architectures and objectives for deformable medical image registration, making the process accessible to non-experts and achieving state-of-the-art results.

Contribution

It proposes a triple-level auto-search framework for jointly optimizing registration network architectures and objectives, reducing expert effort and improving performance.

Findings

AutoReg learns optimal registration networks for diverse datasets.

Achieves state-of-the-art registration accuracy.

Significantly faster computation compared to traditional UNet models.

Abstract

Deformable image registration plays a critical role in various tasks of medical image analysis. A successful registration algorithm, either derived from conventional energy optimization or deep networks requires tremendous efforts from computer experts to well design registration energy or to carefully tune network architectures for the specific type of medical data. To tackle the aforementioned problems, this paper proposes an automated learning registration algorithm (AutoReg) that cooperatively optimizes both architectures and their corresponding training objectives, enabling non-computer experts, e.g., medical/clinical users, to conveniently find off-the-shelf registration algorithms for diverse scenarios. Specifically, we establish a triple-level framework to deduce registration network architectures and objectives with an auto-searching mechanism and cooperating optimization. We conduct image registration experiments on multi-site volume datasets and various registration tasks. Extensive results demonstrate that our AutoReg may automatically learn an optimal deep registration network for given volumes and achieve state-of-the-art performance, also significantly improving computation efficiency than the mainstream UNet architectures (from 0.558 to 0.270 seconds for a 3D image pair on the same configuration).