Encoding Matching Criteria for Cross-domain Deformable Image Registration

TL;DR

This paper introduces a novel registration-oriented encoder that models matching criteria for cross-domain deformable image registration, improving accuracy and adaptability across different medical imaging domains.

Contribution

The paper proposes a new encoder architecture with a general feature encoder and a structural feature encoder, enabling better cross-domain registration and domain adaptation via one-shot learning.

Findings

Effective across three different image domains

Improves registration accuracy and domain adaptability

One-shot learning enhances structural encoder adaptation

Abstract

Most existing deep learning-based registration methods are trained on single-type images to address same-domain tasks.However, cross-domain deformable registration remains challenging.We argue that the tailor-made matching criteria in traditional registration methods is one of the main reason they are applicable in different domains.Motivated by this, we devise a registration-oriented encoder to model the matching criteria of image features and structural features, which is beneficial to boost registration accuracy and adaptability.Specifically, a general feature encoder (Encoder-G) is proposed to capture comprehensive medical image features, while a structural feature encoder (Encoder-S) is designed to encode the structural self-similarity into the global representation.Extensive experiments on images from three different domains prove the efficacy of the proposed method. Moreover, by updating Encoder-S using one-shot learning, our method can effectively adapt to different domains.The code is publicly available at https://github.com/JuliusWang-7/EncoderReg.