Abstract
Glioblastoma is the most common adult primary malignant brain tumor, with grim prognosis and substantial histopathologic and microenvironmental heterogeneity. Here, we present the histopathologic brain tumor sub-region (BraTS) classification challenge, designed in partnership with the clinically authoritative Response Assessment in Neuro-Oncology (RANO) working groups focused on histopathology (RANO-Pathology) and on Artificial Intelligence (AI-RANO) and occurring in conjunction with the 2026 conference of the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society. This challenge describes a large dataset and a community benchmark for the fair evaluation of methodological innovation targeting the automated classification of clinically relevant glioblastoma histopathologic sub-regions in H&E whole-slide images (WSI). Its multi-institutional (n=13) dataset comprises 3,002,093 cases extracted from 368 WSIs (194 patients), partitioned into i) a training cohort (1,631,432 cases) shared with accompanying reference standard annotations, ii) a validation cohort (114,496 cases) shared without annotations, and iii) a hidden hold-out testing cohort (1,256,165 cases). 80 non-annotated WSIs were also provided for training to enable innovations in the domain of self-supervised learning. Reference standard annotations were obtained for 10 distinct histopathologic sub-regions by an international group of 28 neuropathologists. A total of 10 participating teams were evaluated on the hidden hold-out testing cohort, based on the BraTS-Path score, which describes the mean of Matthews correlation coefficient and F1 score. The top-performing method obtained a BraTS-Path score 0.609. This paper summarizes the challenge design, evaluation methods, and the official ranking results. Future work will present a more thorough and extensive post-challenge analysis.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/BraTS_Path_000.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: Not Available
BibTex
@InProceedings{YouSuh_Histopathologic_MICCAISAT2026,
author = { You, Suhang AND Thakur, Siddhesh AND Astaraki, Mehdi AND Toth, Alexander AND Chung, Verena AND Baid, Ujjwal AND LaBella, Dominic AND Correia de Verdier, Maria AND Jiang, Zhifan AND Yordanov, Nikolay AND Menze, Bjoern AND Suero-Molina, Eric AND Cherkezov, Asan AND Kiolbassa, Nora Maren AND Cooper, Lee A. D. AND Ahrendsen, Jared T. AND Ali, Seemaab AND Babaoglu, Berrin AND Balcı, Serdar AND Ballester, Leomar Y. AND Barresi, Valeria AND Fernández Klett, Francisco AND Gubbiotti, Maria A. AND Harmsen, Hannah AND Hortobagyi, Tibor AND Kulac, İbrahim AND López, Giselle Y. AND Lucas, Calixto-Hope G. AND Majeed, Marwan M. AND Miller, Michael L. AND Miletić, Hrvoje AND Nasrallah, MacLean P. AND Phillips, Joanna J. AND Reimann, Regina Rose AND Rodriguez, Michael AND Rodriguez, Fausto J. AND Satgunaseelan, Laveniya AND Schweizer, Leonille AND Stan, Alexandru-Constantin AND Sun, Yu AND Zanazzi, George AND Huang, Raymond Y. AND Farahani, Keyvan AND Aboian, Mariam S. AND Linguraru, Marius George AND Bell, William R. AND Huse, Jason AND Bakas, Spyridon},
title = { { Histopathologic Brain Tumor Sub-region (BraTS) Classification Challenge 2026: Design and Results } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17252},
month = {pending},
page = {pending}
}