EUVIMEDEUVIMED entdecken
Crossref

Prior-guided whole-slide survival prediction with soft histopathologic priors from vision-language foundation models

Songyu Wang, Youjun Chen, Liangqi Zhang, Xiaojun Li, Yuhui Huang, Xiangyu Li iD, Zhihui Feng

Biomedical Physics & Engineering Express · 09.11.2026

DOI: 10.1088/2057-1976/aebfa5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomedical Physics & Engineering Express, online erschienen am 09.11.2026. Verfasst von Songyu Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: College of Information and Management Sciences, Henan Agricultural University, No. 218, Ping'an Avenue, Zhengdong New District, Zhengzhou City, Zheng…. Gefördert von: Henan Province Science and Technology Research Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Whole-slide image (WSI) survival prediction is complicated by heterogeneous tumor morphology, where prognostic evidence is distributed across tumor, stroma, immune infiltration, necrosis, and residual tissue regions and often appears as mixed patch-level tissue evidence. Existing models commonly aggregate patch embeddings directly or convert tissue cues into hard pseudo-labels, which can discard uncertainty at tumor-microenvironment interfaces. We propose a prior-guided WSI survival framework that uses soft histopathologic priors derived from the CONCH vision-language foundation model. Prompt-similarity scores are normalized into patch-level tissue distributions and used to guide probability-weighted tissue-prototype initialization, entropy-weighted prior-consistency learning, and prototype-conditioned graph-context aggregation over a spatial k-nearest-neighbor graph. The residual Others class is retained for representation learning but excluded from prognostic prior alignment. In patient-level 5-fold cross-validation on TCGA KIRC, KIRP, LUAD, STAD, and UCEC cohorts, the model achieved C-index values of 0.7307, 0.8153, 0.6375, 0.6744, and 0.7029, respectively, matching or exceeding the strongest baseline in every cohort. Kaplan-Meier analysis showed significant high-versus low-risk separation across all cohorts, and ablation studies confirmed the contributions of soft priors, entropy weighting, the Others channel, and graph-context aggregation. These results show that retaining patch-level histopathologic uncertainty provides a consistent and interpretable signal for WSI-based survival risk stratification.

Schlagwörter

Stichwörter der Autor:innen

prototype learningsoft histopathologic priorssurvival predictionuncertainty-aware modelingwhole-slide image

Autor:innen und Institutionen

  • College of Information and Management Sciences, Henan Agricultural University, No. 218, Ping'an Avenue, Zhengdong New District, Zhengzhou City, Zhengzhou, 450002, China.
  • School of Mathematics and Statistics, North China University of Water Resources and Electric Power, zhengzhou Henan, Zhengzhou, 450045, China.
  • DeTong Intelligent Technology Co., Ltd, Zhengzhou Henan, Zhengzhou, 450002, China.
  • College of Information and Management Sciences, Henan Agricultural University, Zhengzhou Henan, Zhengzhou, 450002, China.
  • Department of Intelligence and Computing, Tianjin University, No. 135, Yaoguan Road, Jinnan District, Tianjin, Tianjin, 300072, China.

ORCID-Profile: Xiangyu Li

Förderung

  • Henan Province Science and Technology Research Project

Zitieren

APA

Wang, S., Chen, Y., Zhang, L., Li, X., Huang, Y., Li, X., & Feng, Z. (2026). Prior-guided whole-slide survival prediction with soft histopathologic priors from vision-language foundation models. Biomedical Physics & Engineering Express. https://doi.org/10.1088/2057-1976/aebfa5

Vancouver

Wang S, Chen Y, Zhang L, Li X, Huang Y, Li X, et al. Prior-guided whole-slide survival prediction with soft histopathologic priors from vision-language foundation models. Biomedical Physics & Engineering Express. 2026. doi:10.1088/2057-1976/aebfa5

BibTeX
@article{Wang2026,
  author = {Wang, Songyu and Chen, Youjun and Zhang, Liangqi and Li, Xiaojun and Huang, Yuhui and Li, Xiangyu and Feng, Zhihui},
  title = {Prior-guided whole-slide survival prediction with soft histopathologic priors from vision-language foundation models},
  journal = {Biomedical Physics & Engineering Express},
  year = {2026},
  doi = {10.1088/2057-1976/aebfa5}
}
Kostenlos starten →