Inverse Consistency by Construction for Multistep Deep Registration

2023/04/28 by Hastings Greer, Lin Tian, Greer, Hastings +13 · 3 citations
Computer Science · Engineering · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Medical Image Segmentation Techniques #Medical Imaging Techniques and Applications #Medical Imaging and Analysis

paper · pdf · doi:10.48550/arxiv.2305.00087

openalex publication_date 2023/04/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Inverse consistency is a desirable property for image registration. We propose a simple technique to make a neural registration network inverse consistent by construction, as a consequence of its structure, as long as it parameterizes its output transform by a Lie group. We extend this technique to multi-step neural registration by composing many such networks in a way that preserves inverse consistency. This multi-step approach also allows for inverse-consistent coarse to fine registration. We evaluate our technique on synthetic 2-D data and four 3-D medical image registration tasks and obtain excellent registration accuracy while assuring inverse consistency.

Cited by

• Disentangling Progress in Medical Image Registration: Beyond Trend-Driven Architectures towards Domain-Specific Strategies
• Guiding Registration with Emergent Similarity from Pre-Trained Diffusion Models
• Beyond the LUMIR challenge: The pathway to foundational registration models

Related

• A Multiple Decoder CNN for Inverse Consistent 3D Image Registration
• Inverse-Consistent Deep Networks for Unsupervised Deformable Image Registration
• MICS : Multi-steps, Inverse Consistency and Symmetric deep learning registration network
• GradICON: Approximate Diffeomorphisms via Gradient Inverse Consistency
• Towards Saner Deep Image Registration