SMC-UDA: Structure-Modal Constraint for Unsupervised Cross-Domain Renal Segmentation

TL;DR

This paper introduces SMC-UDA, a novel framework for unsupervised cross-domain renal segmentation that leverages edge structure as a domain-invariant feature, improving segmentation accuracy across different imaging modalities.

Contribution

The paper proposes a structure-modal constrained framework that uses edge structure to enhance domain adaptation in renal segmentation, outperforming existing generative UDA methods.

Findings

SMC-UDA outperforms generative UDA methods on renal segmentation tasks.

The framework effectively leverages edge structure for domain-invariant features.

The method demonstrates strong generalization across CT and MRI datasets.

Abstract

Medical image segmentation based on deep learning often fails when deployed on images from a different domain. The domain adaptation methods aim to solve domain-shift challenges, but still face some problems. The transfer learning methods require annotation on the target domain, and the generative unsupervised domain adaptation (UDA) models ignore domain-specific representations, whose generated quality highly restricts segmentation performance. In this study, we propose a novel Structure-Modal Constrained (SMC) UDA framework based on a discriminative paradigm and introduce edge structure as a bridge between domains. The proposed multi-modal learning backbone distills structure information from image texture to distinguish domain-invariant edge structure. With the structure-constrained self-learning and progressive ROI, our methods segment the kidney by locating the 3D spatial structure of the edge. We evaluated SMC-UDA on public renal segmentation datasets, adapting from the labeled source domain (CT) to the unlabeled target domain (CT/MRI). The experiments show that our proposed SMC-UDA has a strong generalization and outperforms generative UDA methods.