Scholarlink

<p>Colorectal cancer (CRC) is a leading malignancy worldwide, where histopathological assessment of hematoxylin and eosin (H&amp;E) stained whole-slide images remains the diagnostic gold standard. However, current computational pathology models suffer from domain shifts, unreliable uncertainty estimation, and spurious correlations, limiting clinical reliability. We present UAD-FM, an Uncertainty-Aware and Causally Adaptive Foundation Model that integrates epistemic-aleatoric uncertainty decomposition, causal test-time adaptation using do-interventions, and post-hoc calibration for trustworthy inference. Across five public CRC datasets (TCGA-COAD/READ, CRAG, DigestPath 2019, NCT-CRC-HE-100K, and LC25000), UAD-FM achieves superior accuracy, calibration, and domain robustness compared with existing foundation models and adaptation baselines. The model also produces interpretable uncertainty maps to support human-AI collaboration. UAD-FM provides a unified, transparent framework for reliable and generalizable CRC pathology diagnosis across heterogeneous clinical settings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Uncertainty-aware and causal test-time adaptive foundation model for robust colorectal cancer pathology diagnosis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月6日

查看全文
  • Shenghan Lou,
  • Genshen Mo,
  • Xiao Zhang,
  • Hao Wang,
  • Hao Li,
  • Keru Ma,
  • Huiying Li,
  • Xinyue Zhang,
  • Meihong Yan,
  • Haonan Xie,
  • Yuze Huang,
  • Chuangqi Li,
  • Siyuan Ma,
  • Hongxue Meng,
  • Lei Cao,
  • Peng Han

摘要

Colorectal cancer (CRC) is a leading malignancy worldwide, where histopathological assessment of hematoxylin and eosin (H&E) stained whole-slide images remains the diagnostic gold standard. However, current computational pathology models suffer from domain shifts, unreliable uncertainty estimation, and spurious correlations, limiting clinical reliability. We present UAD-FM, an Uncertainty-Aware and Causally Adaptive Foundation Model that integrates epistemic-aleatoric uncertainty decomposition, causal test-time adaptation using do-interventions, and post-hoc calibration for trustworthy inference. Across five public CRC datasets (TCGA-COAD/READ, CRAG, DigestPath 2019, NCT-CRC-HE-100K, and LC25000), UAD-FM achieves superior accuracy, calibration, and domain robustness compared with existing foundation models and adaptation baselines. The model also produces interpretable uncertainty maps to support human-AI collaboration. UAD-FM provides a unified, transparent framework for reliable and generalizable CRC pathology diagnosis across heterogeneous clinical settings.