TCSA-UDA: Text-Driven Cross-Semantic Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Maurya, Lalit, Liu, Honghai, Zwiggelaar, Reyer
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911585947090944
author Maurya, Lalit
Liu, Honghai
Zwiggelaar, Reyer
author_facet Maurya, Lalit
Liu, Honghai
Zwiggelaar, Reyer
contents Unsupervised domain adaptation for medical image segmentation remains a significant challenge due to substantial domain shifts across imaging modalities, such as CT and MRI. While recent vision-language representation learning methods have shown promise, their potential in UDA segmentation tasks remains underexplored. To address this gap, we propose TCSA-UDA, a Text-driven Cross-Semantic Alignment framework that leverages domain-invariant textual class descriptions to guide visual representation learning. Our approach introduces a vision-language covariance cosine loss to directly align image encoder features with inter-class textual semantic relations, encouraging semantically meaningful and modality-invariant feature representations. Additionally, we incorporate a prototype alignment module that aligns class-wise pixel-level feature distributions across domains using high-level semantic prototypes. This mitigates residual category-level discrepancies and enhances cross-modal consistency. Extensive experiments on challenging cross-modality cardiac, abdominal, and brain tumor segmentation benchmarks demonstrate that our TCSA-UDA framework significantly reduces domain shift and consistently outperforms state-of-the-art UDA methods, establishing a new paradigm for integrating language-driven semantics into domain-adaptive medical image analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05782
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TCSA-UDA: Text-Driven Cross-Semantic Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation
Maurya, Lalit
Liu, Honghai
Zwiggelaar, Reyer
Computer Vision and Pattern Recognition
Unsupervised domain adaptation for medical image segmentation remains a significant challenge due to substantial domain shifts across imaging modalities, such as CT and MRI. While recent vision-language representation learning methods have shown promise, their potential in UDA segmentation tasks remains underexplored. To address this gap, we propose TCSA-UDA, a Text-driven Cross-Semantic Alignment framework that leverages domain-invariant textual class descriptions to guide visual representation learning. Our approach introduces a vision-language covariance cosine loss to directly align image encoder features with inter-class textual semantic relations, encouraging semantically meaningful and modality-invariant feature representations. Additionally, we incorporate a prototype alignment module that aligns class-wise pixel-level feature distributions across domains using high-level semantic prototypes. This mitigates residual category-level discrepancies and enhances cross-modal consistency. Extensive experiments on challenging cross-modality cardiac, abdominal, and brain tumor segmentation benchmarks demonstrate that our TCSA-UDA framework significantly reduces domain shift and consistently outperforms state-of-the-art UDA methods, establishing a new paradigm for integrating language-driven semantics into domain-adaptive medical image analysis.
title TCSA-UDA: Text-Driven Cross-Semantic Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.05782