TransMedSeg: A Transferable Semantic Framework for Semi-Supervised Medical Image Segmentation

TL;DR

TransMedSeg introduces a transferable semantic framework for semi-supervised medical image segmentation, leveraging domain-invariant features and a novel augmentation module to improve performance across different clinical domains.

Contribution

It proposes a Transferable Semantic Augmentation module that aligns domain-invariant semantics without explicit data generation, enhancing semi-supervised segmentation.

Findings

Outperforms existing semi-supervised methods on medical datasets

Effectively aligns cross-domain semantic features

Implicity enhances feature representations without heavy computation

Abstract

Semi-supervised learning (SSL) has achieved significant progress in medical image segmentation (SSMIS) through effective utilization of limited labeled data. While current SSL methods for medical images predominantly rely on consistency regularization and pseudo-labeling, they often overlook transferable semantic relationships across different clinical domains and imaging modalities. To address this, we propose TransMedSeg, a novel transferable semantic framework for semi-supervised medical image segmentation. Our approach introduces a Transferable Semantic Augmentation (TSA) module, which implicitly enhances feature representations by aligning domain-invariant semantics through cross-domain distribution matching and intra-domain structural preservation. Specifically, TransMedSeg constructs a unified feature space where teacher network features are adaptively augmented towards student network semantics via a lightweight memory module, enabling implicit semantic transformation without explicit data generation. Interestingly, this augmentation is implicitly realized through an expected transferable cross-entropy loss computed over the augmented teacher distribution. An upper bound of the expected loss is theoretically derived and minimized during training, incurring negligible computational overhead. Extensive experiments on medical image datasets demonstrate that TransMedSeg outperforms existing semi-supervised methods, establishing a new direction for transferable representation learning in medical image analysis.