Browse by author
Lookup NU author(s): Dr Jess Dyson
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
© 2026 The Authors. Digital pathology enables large multi-center studies of histological specimens, but differences in staining protocols and slide quality can compromise the comparability of quantitative results. We analyzed 686 PicroSirius Red-stained liver biopsies from 4 independent cohorts spanning more than 20 clinical sites to assess how stain variability affects automated fibrosis quantification and model uncertainty. An U-Net ensemble was trained to segment collagen and to estimate pixel- and tile-level predictive uncertainty. Across markedly heterogeneous staining conditions, the ensemble achieved strong segmentation performance (Dice 0.83–0.90) and produced informative uncertainty maps that identified artifacts and out-of-distribution regions. Epistemic uncertainty values were typically below 0.002, providing a practical criterion for flagging unreliable predictions. Our results demonstrate that ensemble-based uncertainty estimation complements stain-standardization efforts by quantifying prediction confidence directly from model outputs, improving the reliability and interpretability of collagen proportionate-area measurements across multi-center datasets. This framework supports more trustworthy and reproducible digital-pathology workflows for fibrosis assessment and other histological applications.
Author(s): Wojciechowska M, Malacrino S, Windell D, Culver EL, Dyson JK, Rittscher J
Publication type: Article
Publication status: Published
Journal: Journal of Pathology Informatics
Year: 2026
Volume: 21
Print publication date: 01/04/2026
Online publication date: 20/03/2026
Acceptance date: 18/03/2026
Date deposited: 06/05/2026
ISSN (print): 2229-5089
ISSN (electronic): 2153-3539
Publisher: Elsevier BV
URL: https://doi.org/10.1016/j.jpi.2026.100653
DOI: 10.1016/j.jpi.2026.100653
Data Access Statement: The source code for the slide color analysis tool is available under appendix.
Altmetrics provided by Altmetric
