EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Semi-SwinUNeTR: Towards 3D Swin Vision Transformer-Based UNet for Medical Image Segmentation with Limited Annotations

Yinbing Tian iD, Ziyang Wang iD, Li Guo iD

Bioengineering · 17.06.2026 · 13 (6) · 695

DOI: 10.3390/bioengineering13060695 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 17.06.2026. Verfasst von Yinbing Tian und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Beijing Natural Science Foundation, Key Research and Development Program of Hainan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate brain tumor segmentation from magnetic resonance imaging (MRI) is essential for computer-assisted diagnosis, treatment planning, and disease monitoring. However, brain tumors usually exhibit irregular, heterogeneous, and multi-scale spatial patterns with complex and ambiguous boundaries. At the same time, the performance of deep segmentation models is often constrained by the limited availability of voxel-level annotations, which are expensive and time-consuming to obtain. To address these challenges, this paper proposes Semi-SwinUNeTR, a semi-supervised framework for 3D brain tumor segmentation with limited annotated data. The proposed method adopts SwinUNeTR as the segmentation backbone, enabling hierarchical volumetric representation learning through shifted-window self-attention while preserving the encoder–decoder structure required for dense prediction. On top of this backbone, we introduce a dual-consistency semi-supervised learning strategy, consisting of mean teacher-based model consistency and interpolation consistency-based data consistency. In addition, voxel-wise consistency weights are used to redistribute semi-supervised supervision toward structurally complex and boundary-irregular tumor regions without changing the SwinUNeTR backbone. Experiments on the BraTS 2019 benchmark demonstrate that the proposed framework achieves strong performance across different annotation ratios. The original Semi-SwinUNeTR achieves Dice scores of 84.93%, 86.25%, 87.05%, and 87.83% under the 10%, 20%, 40%, and 80% labeled-data settings, respectively. With the weighted consistency extension, the Dice scores are further improved to 85.64%, 87.94%, and 88.59% under the 10%, 20%, and 80% labeled-data settings, respectively, while the corresponding HD95 values are reduced to 8.9826, 8.1854, and 7.4533. These results indicate that combining a SwinUNeTR backbone with complementary model consistency, data consistency, and voxel-wise consistency weighting is an effective strategy for semi-supervised volumetric medical image segmentation under limited annotation.

Schlagwörter

Stichwörter der Autor:innen

SwinUNeTRauxiliary weightingbiomedical image analysisinterpolation consistencysemi-supervised learningvision transformervolumetric medical image segmentation

Autor:innen und Institutionen

  • School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • Engineering Research Center of Blockchain and Network Convergence Technology, Ministry of Education, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • National Engineering Research Center for Mobile Internet Security Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • Key Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, Beijing 100876, China.
  • School of Computer Science and Digital Technologies, Aston University, Birmingham B4 7ET, UK.

ORCID-Profile: Yinbing Tian, Ziyang Wang, Li Guo

Förderung

  • Beijing Natural Science Foundation
  • Key Research and Development Program of Hainan Province

Zitieren

APA

Tian, Y., Wang, Z., & Guo, L. (2026). Semi-SwinUNeTR: Towards 3D Swin Vision Transformer-Based UNet for Medical Image Segmentation with Limited Annotations. Bioengineering, 13(6), 695. https://doi.org/10.3390/bioengineering13060695

Vancouver

Tian Y, Wang Z, Guo L. Semi-SwinUNeTR: Towards 3D Swin Vision Transformer-Based UNet for Medical Image Segmentation with Limited Annotations. Bioengineering. 2026;13(6):695. doi:10.3390/bioengineering13060695

BibTeX
@article{Tian2026,
  author = {Tian, Yinbing and Wang, Ziyang and Guo, Li},
  title = {Semi-SwinUNeTR: Towards 3D Swin Vision Transformer-Based UNet for Medical Image Segmentation with Limited Annotations},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {6},
  pages = {695},
  doi = {10.3390/bioengineering13060695}
}
Kostenlos starten →