Region Specific Optimization (RSO)-based Deep Interactive Registration

TL;DR

This paper introduces a region-specific optimization method for deep learning-based medical image registration, improving accuracy in challenging regions through interactive, region-specific adjustments during the registration process.

Contribution

It proposes a novel two-level test time optimization approach, including region-specific optimization, to enhance registration accuracy in difficult regions, with an efficient three-step workflow.

Findings

Outperforms conventional registration methods quantitatively and qualitatively.

Effectively improves registration accuracy in challenging regions.

Enables interactive, clinician-guided region-specific optimization.

Abstract

Medical image registration is a fundamental and vital task which will affect the efficacy of many downstream clinical tasks. Deep learning (DL)-based deformable image registration (DIR) methods have been investigated, showing state-of-the-art performance. A test time optimization (TTO) technique was proposed to further improve the DL models' performance. Despite the substantial accuracy improvement with this TTO technique, there still remained some regions that exhibited large registration errors even after many TTO iterations. To mitigate this challenge, we firstly identified the reason why the TTO technique was slow, or even failed, to improve those regions' registration results. We then proposed a two-levels TTO technique, i.e., image-specific optimization (ISO) and region-specific optimization (RSO), where the region can be interactively indicated by the clinician during the registration result reviewing process. For both efficiency and accuracy, we further envisioned a three-step DL-based image registration workflow. Experimental results showed that our proposed method outperformed the conventional method qualitatively and quantitatively.