Training Beyond Convergence: Grokking nnU-Net for Glioma Segmentation in Sub-Saharan MRI
Barakat M, Salah O, Yasser A, Ahmed M, Arief Z, Khan W, Zhang D, Iorumbur A, Raymond C, Barakat M, Magdy N (2026)
Publication Type: Conference contribution
Publication year: 2026
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 16376 LNCS
Pages Range: 307-317
Conference Proceedings Title: Lecture Notes in Computer Science
Event location: Daejeon
ISBN: 9783032163646
DOI: 10.1007/978-3-032-16365-3_28
Abstract
Gliomas are placing an increasingly clinical burden on Sub-Saharan Africa (SSA). In the region, the median survival for patients remains under two years, and access to diagnostic imaging is extremely limited. These constraints highlight an urgent need for automated tools that can extract the maximum possible information from each available scan, tools that are specifically trained on local data, rather than adapted from high-income settings where conditions are vastly different. We utilize the Brain Tumor Segmentation (BraTS) Africa 2025 Challenge dataset, an expert annotated collection of glioma MRIs. Our objectives are: (i) establish a strong baseline with nnUNet on this dataset, and (ii) explore whether the celebrated “grokking” phenomenon an abrupt, late training jump from memorization to superior generalization can be triggered to push performance without extra labels. We evaluate two training regimes. The first is a fast, budget-conscious approach that limits optimization to just a few epochs, reflecting the constrained GPU resources typically available in African institutions. Despite this limitation, nnUNet achieves strong Dice scores: 92.3% for whole tumor (WH), 86.6% for tumor core (TC), and 86.3% for enhancing tumor (ET). The second regime extends training well beyond the point of convergence, aiming to trigger a grokking-driven performance leap. With this approach, we were able to achieve grokking and enhanced our results to higher Dice scores: 92.2% for whole tumor (WH), 90.1% for tumor core (TC), and 90.2% for enhancing tumor (ET).
Involved external institutions
University of British Columbia
Canada (CA)
Federal University of Technology Minna (FUTMINNA)
Nigeria (NG)
Montreal Neurological Institute and Hospital (MNI, The Neuro)
Canada (CA)
Multimedia University (MMU)
Malaysia (MY)
Canada (CA)
Federal University of Technology Minna (FUTMINNA)
Nigeria (NG)
Montreal Neurological Institute and Hospital (MNI, The Neuro)
Canada (CA)
Multimedia University (MMU)
Malaysia (MY)
How to cite
APA:
Barakat, M., Salah, O., Yasser, A., Ahmed, M., Arief, Z., Khan, W.,... Magdy, N. (2026). Training Beyond Convergence: Grokking nnU-Net for Glioma Segmentation in Sub-Saharan MRI. In Spyridon Bakas, Emily Dennis, Mehdi Astaraki, Ujjwal Baid, Gian Marco Conte, Martha Foltyn-Dumitru, Zhifan Jiang, Marius George Linguraru, Dominic Labella, Marie-Christin Metz, Udunna Anazodo, Maria Correia de Verdier, Florian Kofler, Hongwei Bran Li, Nazanin Maleki (Eds.), Lecture Notes in Computer Science (pp. 307-317). Daejeon, KR: Springer Science and Business Media Deutschland GmbH.
MLA:
Barakat, Mohtady, et al. "Training Beyond Convergence: Grokking nnU-Net for Glioma Segmentation in Sub-Saharan MRI." Proceedings of the Brain TumorS Lighthouse Cluster of Challenges, and the Automated Identification of Moderate-Severe Traumatic Brain Injury Lesions Challenge, BraTS 2025 and AIMS-TBI 2025, held in Conjunction International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2025, Daejeon Ed. Spyridon Bakas, Emily Dennis, Mehdi Astaraki, Ujjwal Baid, Gian Marco Conte, Martha Foltyn-Dumitru, Zhifan Jiang, Marius George Linguraru, Dominic Labella, Marie-Christin Metz, Udunna Anazodo, Maria Correia de Verdier, Florian Kofler, Hongwei Bran Li, Nazanin Maleki, Springer Science and Business Media Deutschland GmbH, 2026. 307-317.
BibTeX: Download