EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Delving into Unreliable Pseudo-Labels for Semi-Supervised Medical Image Segmentation via Conformal Selection

Jialin Shi, Zongyao Yang, Youquan Yang, Kai Wu, Zongjie Wang

Electronics · 2026-02-20 · 15 (4) · 886

DOI: 10.3390/electronics15040886 ↗

At a glance

Journal article in Electronics, published online on 2026-02-20. Written by Jialin Shi and 4 co-authors. Cited 1 times so far (Crossref); 32 references. The full text is freely available (open access, CC BY 4.0). Funded by: Key Research and Development Program of NingXia, Noncommunicable Chronic Diseases-National Science and Technology Major Project, Guangdong Basic and Applied Basic Research Foundation.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Semi-supervised medical image segmentation has recently achieved great success, but assigning trustworthy pseudo-labels to unlabeled images has been a difficult problem in medical image processing. A common solution is to select reliable predicted pixels as the pseudo-labels. However, unreliable pixels are often concentrated in the edge areas of the foreground and background in medical tasks. Directly discarding these pixels will result in this important information never being available. The foreground of medical images is usually surrounded by the edge area. This section of pixels is a mixture of the two categories, which makes it very difficult to distinguish. To address these problems, we propose a semi-supervised medical segmentation framework that combines conformal prediction and contrastive learning. Our framework can use conformal prediction to select pseudo-labels with high confidence and preserve important boundary information. Furthermore, the segmentation performance of edge regions can be improved using contrastive learning between edge categories and non-edge categories. Extensive experiments on multiple benchmarks show that our framework consistently outperforms state-of-the-art methods.

Funding

  • Key Research and Development Program of NingXia
  • Noncommunicable Chronic Diseases-National Science and Technology Major Project
  • Guangdong Basic and Applied Basic Research Foundation

Cite

APA

Shi, J., Yang, Z., Yang, Y., Wu, K., & Wang, Z. (2026). Delving into Unreliable Pseudo-Labels for Semi-Supervised Medical Image Segmentation via Conformal Selection. Electronics, 15(4), 886. https://doi.org/10.3390/electronics15040886

Vancouver

Shi J, Yang Z, Yang Y, Wu K, Wang Z. Delving into Unreliable Pseudo-Labels for Semi-Supervised Medical Image Segmentation via Conformal Selection. Electronics. 2026;15(4):886. doi:10.3390/electronics15040886

BibTeX
@article{Shi2026,
  author = {Shi, Jialin and Yang, Zongyao and Yang, Youquan and Wu, Kai and Wang, Zongjie},
  title = {Delving into Unreliable Pseudo-Labels for Semi-Supervised Medical Image Segmentation via Conformal Selection},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {4},
  pages = {886},
  doi = {10.3390/electronics15040886}
}
Start for free →