SAMConvex: Fast Discrete Optimization for CT Registration using Self-supervised Anatomical Embedding and Correlation Pyramid

TL;DR

SAMConvex introduces a fast, coarse-to-fine discrete optimization approach for CT registration that leverages self-supervised anatomical embeddings to efficiently capture both local and global features, outperforming existing methods.

Contribution

It proposes a novel, rapid optimization method using self-supervised embeddings and correlation pyramids for improved CT image registration.

Findings

Outperforms state-of-the-art methods on multiple CT datasets.

Achieves registration in approximately 2 seconds per image pair.

Effective for large transformations due to global feature representation.

Abstract

Estimating displacement vector field via a cost volume computed in the feature space has shown great success in image registration, but it suffers excessive computation burdens. Moreover, existing feature descriptors only extract local features incapable of representing the global semantic information, which is especially important for solving large transformations. To address the discussed issues, we propose SAMConvex, a fast coarse-to-fine discrete optimization method for CT registration that includes a decoupled convex optimization procedure to obtain deformation fields based on a self-supervised anatomical embedding (SAM) feature extractor that captures both local and global information. To be specific, SAMConvex extracts per-voxel features and builds 6D correlation volumes based on SAM features, and iteratively updates a flow field by performing lookups on the correlation volumes with a coarse-to-fine scheme. SAMConvex outperforms the state-of-the-art learning-based methods and optimization-based methods over two inter-patient registration datasets (Abdomen CT and HeadNeck CT) and one intra-patient registration dataset (Lung CT). Moreover, as an optimization-based method, SAMConvex only takes $\sim2$s ($\sim5s$ with instance optimization) for one paired images.