Automated semantic segmentation of whole-slide images (WSIs) stained with hematoxylin and eosin (H&E) is essential for large-scale artificial intelligence-based biomarker analysis in breast cancer. However, existing public datasets for breast cancer segmentation lack the morphological diversity needed to support model generalizability and robust biomarker validation across heterogeneous patient cohorts. We introduce BrEast cancEr hisTopathoLogy sEgmentation (BEETLE), a dataset for multiclass semantic segmentation of H&E-stained breast cancer WSIs. It consists of 587 biopsies and resections from three collaborating clinical centers and two public datasets, digitized using seven scanners, and covers all molecular subtypes and histological grades. Using diverse annotation strategies, we collected annotations across four classes - invasive epithelium, non-invasive epithelium, necrosis, and other - with particular focus on morphologies underrepresented in existing datasets, such as ductal carcinoma in situ and dispersed lobular tumor cells. The dataset's diversity and relevance to the rapidly growing field of automated biomarker quantification in breast cancer ensure its high potential for reuse. Finally, we provide a well-curated, multicentric external evaluation set to enable standardized benchmarking of breast cancer segmentation models.
@article{arxiv.2510.02037,
title = {A Multicentric Dataset for Training and Benchmarking Breast Cancer Segmentation in H&E Slides},
author = {Carlijn Lems and Leslie Tessier and John-Melle Bokhorst and Mart van Rijthoven and Witali Aswolinskiy and Matteo Pozzi and Natalie Klubickova and Suzanne Dintzis and Michela Campora and Maschenka Balkenhol and Peter Bult and Joey Spronck and Thomas Detone and Mattia Barbareschi and Enrico Munari and Giuseppe Bogina and Jelle Wesseling and Esther H. Lips and Francesco Ciompi and Frédérique Meeuwsen and Jeroen van der Laak},
journal= {arXiv preprint arXiv:2510.02037},
year = {2025}
}
Comments
Our dataset is available at https://zenodo.org/records/16812932 , our code is available at https://github.com/DIAGNijmegen/beetle , and our benchmark is available at https://beetle.grand-challenge.org/