Scholarlink

<p>Artificial intelligence applications in liver pathology remain limited, with existing tools either narrowly focused or lacking external validation. This study introduces HOTSPoT, an open-source, validated transformer-based model for automated segmentation of portal tracts in H&amp;E-stained liver biopsy whole slide images. A multi-institutional dataset of 223 cases was used, with annotations by expert hepatopathologists. HOTSPoT achieved high performance with mean Dice scores of 0.92 (train/val) and 0.91 (test), and mean IoUs of 0.86, 0.85, and 0.84, respectively, showing minimal domain shift. Automated portal tract quantification showed strong concordance with manual assessments (κ up to 0.90), and portal area correlated with fibrosis stage (<i>r</i> = 0.87, <i>p</i> &lt; 0.001). The model is available as a TorchScript file with a modified WSInfer library, enabling efficient WSI-level inference and integration with QuPath for advanced pathology analysis.</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

Automating liver biopsy segmentation with a robust, open-source tool for pathology research: the HOTSPoT model

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月18日

查看全文
  • Giorgio Cazzaniga,
  • Vincenzo L’Imperio,
  • Emanuela Bonoldi,
  • María-Carlota Londoño,
  • João Madaleno,
  • Augusta Cipriano,
  • Tom J. G. Gevers,
  • Iryna V. Samarska,
  • Özgür M. Koc,
  • Alejandra Villamil,
  • Maria Florencia Sanchez,
  • Vincenza Calvaruso,
  • Alberto Quattrocchi,
  • Daniela Cabibi,
  • Bastian Engel,
  • Federica Malinverno,
  • Elisa Merelli,
  • Laura Cristoferi,
  • Marco Carbone,
  • Fabio Pagni,
  • Pietro Invernizzi,
  • Alessio Gerussi

摘要

Artificial intelligence applications in liver pathology remain limited, with existing tools either narrowly focused or lacking external validation. This study introduces HOTSPoT, an open-source, validated transformer-based model for automated segmentation of portal tracts in H&E-stained liver biopsy whole slide images. A multi-institutional dataset of 223 cases was used, with annotations by expert hepatopathologists. HOTSPoT achieved high performance with mean Dice scores of 0.92 (train/val) and 0.91 (test), and mean IoUs of 0.86, 0.85, and 0.84, respectively, showing minimal domain shift. Automated portal tract quantification showed strong concordance with manual assessments (κ up to 0.90), and portal area correlated with fibrosis stage (r = 0.87, p < 0.001). The model is available as a TorchScript file with a modified WSInfer library, enabling efficient WSI-level inference and integration with QuPath for advanced pathology analysis.