Domain-invariant Mixed-domain Semi-supervised Medical Image Segmentation with Clustered Maximum Mean Discrepancy Alignment

TL;DR

This paper introduces a novel semi-supervised medical image segmentation method that effectively handles mixed-domain data with unknown domain labels by combining data augmentation and domain-invariant feature alignment.

Contribution

It proposes a domain-invariant framework using Copy-Paste augmentation and Cluster MMD alignment within a teacher-student model for improved segmentation across multiple unknown domains.

Findings

Outperforms existing semi-supervised and domain adaptation methods on Fundus and M&Ms benchmarks.

Achieves robust segmentation with very few labeled examples across mixed domains.

Demonstrates effective domain-invariant feature learning in complex real-world scenarios.

Abstract

Deep learning has shown remarkable progress in medical image semantic segmentation, yet its success heavily depends on large-scale expert annotations and consistent data distributions. In practice, annotations are scarce, and images are collected from multiple scanners or centers, leading to mixed-domain settings with unknown domain labels and severe domain gaps. Existing semi-supervised or domain adaptation approaches typically assume either a single domain shift or access to explicit domain indices, which rarely hold in real-world deployment. In this paper, we propose a domain-invariant mixed-domain semi-supervised segmentation framework that jointly enhances data diversity and mitigates domain bias. A Copy-Paste Mechanism (CPM) augments the training set by transferring informative regions across domains, while a Cluster Maximum Mean Discrepancy (CMMD) block clusters unlabeled features and aligns them with labeled anchors via an MMD objective, encouraging domain-invariant representations. Integrated within a teacher-student framework, our method achieves robust and precise segmentation even with very few labeled examples and multiple unknown domain discrepancies. Experiments on Fundus and M&Ms benchmarks demonstrate that our approach consistently surpasses semi-supervised and domain adaptation methods, establishing a potential solution for mixed-domain semi-supervised medical image segmentation.