# One registration is worth two segmentations

Source: [One registration is worth two segmentations](<https://www.sota2.com/research/paper/68fa734b1e209131cea5b9d9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:48:58.476Z

Authors: Shiqi Huang, Tingfa Xu, Ziyi Shen, Shaheer Ullah Saeed, Wen Yan, Dean Barratt, Yipeng Hu

Published: 2024-05-17T16:14:32.000Z

## Abstract

The goal of image registration is to establish spatial correspondence between two or more images, traditionally through dense displacement fields (DDFs) or parametric transformations (e.g., rigid, affine, and splines). Rethinking the existing paradigms of achieving alignment via spatial transformations, we uncover an alternative but more intuitive correspondence representation: a set of corresponding regions-of-interest (ROI) pairs, which we demonstrate to have sufficient representational capability as other correspondence representation methods.Further, it is neither necessary nor sufficient for these ROIs to hold specific anatomical or semantic significance. In turn, we formulate image registration as searching for the same set of corresponding ROIs from both moving and fixed images - in other words, two multi-class segmentation tasks on a pair of images. For a general-purpose and practical implementation, we integrate the segment anything model (SAM) into our proposed algorithms, resulting in a SAM-enabled registration (SAMReg) that does not require any training data, gradient-based fine-tuning or engineered prompts. We experimentally show that the proposed SAMReg is capable of segmenting and matching multiple ROI pairs, which establish sufficiently accurate correspondences, in three clinical applications of registering prostate MR, cardiac MR and abdominal CT images. Based on metrics including Dice and target registration errors on anatomical structures, the proposed registration outperforms both intensity-based iterative algorithms and DDF-predicting learning-based networks, even yielding competitive performance with weakly-supervised registration which requires fully-segmented training data.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Medical Image Registration on MR-Cardiac (intra-subject)](<https://www.sota2.com/research/sota/medical-image-registration-on-mr-cardiac-intra-subject.md>): Medical Image Registration; MR-Cardiac (intra-subject); Dice Score: 80.28.

- [Image Registration on Abdomen CT (inter-subject)](<https://www.sota2.com/research/sota/image-registration-on-abdomen-ct-inter-subject.md>): Image Registration; Abdomen CT (inter-subject); Dice Coefficient: 7174.

- [Medical Image Registration on MR-Prostate (inter-subject)](<https://www.sota2.com/research/sota/medical-image-registration-on-mr-prostate-inter-subject.md>): Medical Image Registration; MR-Prostate (inter-subject); Dice: 0.7567.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2405.10879>)

[PDF](<https://arxiv.org/pdf/2405.10879.pdf>)

[Implementation](<https://github.com/sqhuang0103/SAMReg.git>)
