Towards Saner Deep Image Registration

TL;DR

This paper introduces a regularization-based method to improve the reliability and discriminative behavior of deep image registration models, especially for medical imaging, without compromising their performance.

Contribution

It proposes a novel sanity-enforcer regularization technique that enforces inverse consistency and discrimination, supported by theoretical guarantees and experimental validation.

Findings

Reduces inverse consistency errors in deep registration models

Increases discriminative power of registration models

Maintains performance while improving model sanity

Abstract

With recent advances in computing hardware and surges of deep-learning architectures, learning-based deep image registration methods have surpassed their traditional counterparts, in terms of metric performance and inference time. However, these methods focus on improving performance measurements such as Dice, resulting in less attention given to model behaviors that are equally desirable for registrations, especially for medical imaging. This paper investigates these behaviors for popular learning-based deep registrations under a sanity-checking microscope. We find that most existing registrations suffer from low inverse consistency and nondiscrimination of identical pairs due to overly optimized image similarities. To rectify these behaviors, we propose a novel regularization-based sanity-enforcer method that imposes two sanity checks on the deep model to reduce its inverse consistency errors and increase its discriminative power simultaneously. Moreover, we derive a set of theoretical guarantees for our sanity-checked image registration method, with experimental results supporting our theoretical findings and their effectiveness in increasing the sanity of models without sacrificing any performance. Our code and models are available at https://github.com/tuffr5/Saner-deep-registration.