Fully automatic HER2 tissue segmentation for interpretable HER2 scoring

Öttl M, Steenpaß J, Wilm F, Qiu J, Rübner M, Lang-Schwarz C, Taverna C, Tava F, Hartmann A, Hübner H, Beckmann M, Fasching P, Maier A, Erber R, Breininger K (2025)

Publication Type: Journal article

Publication year: 2025

Journal

Journal of Pathology Informatics Medknow Publications

Book Volume: 17

Article Number: 100435

DOI: 10.1016/j.jpi.2025.100435

Abstract

Breast cancer is the most common cancer in women, with HER2 (human epidermal growth factor receptor 2) overexpression playing a critical role in regulating cell growth and division. HER2 status, assessed according to established scoring guidelines, offers important information for treatment selection. However, the complexity of the task leads to variability in human rater assessments. In this work, we propose a fully automated, interpretable HER2 scoring pipeline based on pixel-level semantic segmentations, designed to align with clinical guidelines. Using polygon annotations, our method balances annotation effort with the ability to capture fine-grained details and larger structures, such as non-invasive tumor tissue. To enhance HER2 segmentation, we propose the use of a Wasserstein Dice loss to model class relationships, ensuring robust segmentation and HER2 scoring performance. Additionally, based on observations of pathologists' behavior in clinical practice, we propose a calibration step to the scoring rules, which positively impacts the accuracy and consistency of automated HER2 scoring. Our approach achieves an F1 score of 0.832 on HER2 scoring, demonstrating its effectiveness. This work establishes a potent segmentation pipeline that can be further leveraged to analyze HER2 expression in breast cancer tissue.

Authors with CRIS profile

Mathias Öttl Lehrstuhl für Informatik 5 (Mustererkennung) Jana Steenpaß Lehrstuhl für Allgemeine Pathologie und Pathologische Anatomie Frauke Wilm Lehrstuhl für Informatik 5 (Mustererkennung) Jingna Qiu Department Artificial Intelligence in Biomedical Engineering (AIBE) Matthias Rübner Department of Obstetrics and Gynaecology Corinna Lang-Schwarz
Arndt Hartmann Lehrstuhl für Allgemeine Pathologie und Pathologische Anatomie Hanna Hübner Department of Obstetrics and Gynaecology Matthias Beckmann Lehrstuhl für Geburtshilfe und Frauenheilkunde Peter Fasching Professur für Translationale Frauenheilkunde und Geburtshilfe Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung) Ramona Erber Institute of Pathology Katharina Breininger Department Artificial Intelligence in Biomedical Engineering (AIBE)

Involved external institutions

Klinikum Bayreuth DE Germany (DE)

How to cite

APA:

Öttl, M., Steenpaß, J., Wilm, F., Qiu, J., Rübner, M., Lang-Schwarz, C.,... Breininger, K. (2025). Fully automatic HER2 tissue segmentation for interpretable HER2 scoring. Journal of Pathology Informatics, 17. https://doi.org/10.1016/j.jpi.2025.100435

MLA:

Öttl, Mathias, et al. "Fully automatic HER2 tissue segmentation for interpretable HER2 scoring." Journal of Pathology Informatics 17 (2025).

BibTeX: Download