Self-distillation double student network for semi-supervised medical image segmentation
Frontiers in Physiology · 29.05.2026 · 17
DOI: 10.3389/fphys.2026.1755067 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Physiology, online erschienen am 29.05.2026. Verfasst von Huaxiang Liu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Taizhou University Affiliated Taizhou Central Hospital, Taizhou University, Taizhou, Zhejiang, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Semi-supervised learning (SSL) has garnered considerable attention in medical image segmentation due to its ability to leverage abundant unlabeled data, thereby significantly alleviating the dependency on expensive pixel-wise annotations. However, existing Mean-Teacher-based frameworks typically guide the student network using a teacher model in a unidirectional or loosely interactive manner, which fails to fully exploit the complementary relationships between different networks and often struggles with low-confidence predictions that degrade boundary accuracy. To address these limitations, we propose SDBS-Net, a novel semi-supervised dual-student self-distillation network for 3D medical image segmentation. The architecture comprises a shared encoder and two parallel decoders, forming two networks (Model I and Model II) that process both labeled and unlabeled volumes simultaneously. To enable effective bidirectional knowledge transfer, we introduce two synergistic training schemas: a prior knowledge learning (PKL) schema with hierarchical difference (HD) modules and a Self-Distillation Learning (SDL) schema. The PKL schema extracts and inject multi-scale discrepancy features from Model I into Model II, enhancing discriminative representation and boundary delineation, while the SDL schema employs multi-level soft-label and feature distillation losses to correct confirmation bias and refine overlooked regions in a mutually reinforcing manner. Extensive experiments on two challenging benchmarks such as the Left Atrium (LA) dataset and the Pancreas-CT dataset demonstrate the superiority of SDBS-Net. With only 10% and 20% labeled data, our method achieves Dice scores of 88.19% and 90.51% on the LA dataset, and 71.39% and 79.37% on the Pancreas-CT dataset, respectively, consistently outperforming state-of-the-art semi-supervised approaches while approaching fully supervised performance.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Taizhou University Affiliated Taizhou Central Hospital, Taizhou University, Taizhou, Zhejiang, China.
- Artificial Intelligence School, Taizhou University, Taizhou, Zhejiang, China.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Liu, H., Li, X., Jin, J., Zhang, S., Fu, Y., Fang, J., & Wang, G. (2026). Self-distillation double student network for semi-supervised medical image segmentation. Frontiers in Physiology, 17. https://doi.org/10.3389/fphys.2026.1755067
Vancouver
Liu H, Li X, Jin J, Zhang S, Fu Y, Fang J, et al. Self-distillation double student network for semi-supervised medical image segmentation. Frontiers in Physiology. 2026;17. doi:10.3389/fphys.2026.1755067
BibTeX
@article{Liu2026,
author = {Liu, Huaxiang and Li, Xin and Jin, Jie and Zhang, Shiqing and Fu, Youyao and Fang, Jiangxiong and Wang, Guoyu},
title = {Self-distillation double student network for semi-supervised medical image segmentation},
journal = {Frontiers in Physiology},
year = {2026},
volume = {17},
doi = {10.3389/fphys.2026.1755067}
}
EUVIMED