{"ID":2848419,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.26826","arxiv_id":"2510.26826","title":"UP2D: Uncertainty-aware Progressive Pseudo-label Denoising for Source-Free Domain Adaptive Medical Image Segmentation","abstract":"Medical image segmentation models face severe performance drops under domain shifts, especially when data sharing constraints prevent access to source images. We present a novel Uncertainty-aware Progressive Pseudo-label Denoising (UP2D) framework for source-free domain adaptation (SFDA), designed to mitigate noisy pseudo-labels and class imbalance during adaptation. UP2D integrates three key components: (i) a Refined Prototype Filtering module that suppresses uninformative regions and constructs reliable class prototypes to denoise pseudo-labels, (ii) an Uncertainty-Guided EMA (UG-EMA) strategy that selectively updates the teacher model based on spatially weighted boundary uncertainty, and (iii) a quantile-based entropy minimization scheme that focuses learning on ambiguous regions while avoiding overconfidence on easy pixels. This single-stage student-teacher framework progressively improves pseudo-label quality and reduces confirmation bias. Extensive experiments on three challenging retinal fundus benchmarks demonstrate that UP2D achieves state-of-the-art performance across both standard and open-domain settings, outperforming prior UDA and SFDA approaches while maintaining superior boundary precision.","short_abstract":"Medical image segmentation models face severe performance drops under domain shifts, especially when data sharing constraints prevent access to source images. We present a novel Uncertainty-aware Progressive Pseudo-label Denoising (UP2D) framework for source-free domain adaptation (SFDA), designed to mitigate noisy pse...","url_abs":"https://arxiv.org/abs/2510.26826","url_pdf":"https://arxiv.org/pdf/2510.26826v1","authors":"[\"Quang-Khai Bui-Tran\",\"Thanh-Huy Nguyen\",\"Manh D. Ho\",\"Thinh B. Lam\",\"Vi Vu\",\"Hoang-Thien Nguyen\",\"Phat Huynh\",\"Ulas Bagci\"]","published":"2025-10-29T06:43:12Z","proceeding":"eess.IV","tasks":"[\"eess.IV\"]","methods":"[]","has_code":false}