TY - GEN
T1 - Federated Fine-Tuning of SAM-Med3D for MRI-Based Dementia Classification
AU - Mouheb, Kaouther
AU - Elbatel, Marawan
AU - Papma, Janne
AU - Biessels, Geert Jan
AU - Claassen, Jurgen
AU - Middelkoop, Huub
AU - van Munster, Barbara
AU - van der Flier, Wiesje
AU - Ramakers, Inez
AU - Klein, Stefan
AU - Bron, Esther E.
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - While foundation models (FMs) offer strong potential for AI-based dementia diagnosis, their integration into federated learning (FL) systems remains underexplored. In this benchmarking study, we systematically evaluate the impact of key design choices: classification head architecture, fine-tuning strategy, and aggregation method, on the performance and efficiency of federated FM tuning using brain MRI data. Using a large multi-cohort dataset, we find that the architecture of the classification head substantially influences performance, freezing the FM encoder achieves comparable results to full fine-tuning, and advanced aggregation methods outperform standard federated averaging. Our results offer practical insights for deploying FMs in decentralized clinical settings and highlight trade-offs that should guide future method development.
AB - While foundation models (FMs) offer strong potential for AI-based dementia diagnosis, their integration into federated learning (FL) systems remains underexplored. In this benchmarking study, we systematically evaluate the impact of key design choices: classification head architecture, fine-tuning strategy, and aggregation method, on the performance and efficiency of federated FM tuning using brain MRI data. Using a large multi-cohort dataset, we find that the architecture of the classification head substantially influences performance, freezing the FM encoder achieves comparable results to full fine-tuning, and advanced aggregation methods outperform standard federated averaging. Our results offer practical insights for deploying FMs in decentralized clinical settings and highlight trade-offs that should guide future method development.
KW - Dementia
KW - Federated learning
KW - Foundation models
KW - MRI
UR - https://www.scopus.com/pages/publications/105018308033
U2 - 10.1007/978-3-032-05663-4_7
DO - 10.1007/978-3-032-05663-4_7
M3 - Conference contribution
SN - 9783032056658
VL - 16135 LNCS
T3 - Lecture Notes in Computer Science
SP - 69
EP - 79
BT - Bridging Regulatory Science and Medical Imaging Evaluation; and Distributed, Collaborative, and Federated Learning - 1st International Workshop, BRIDGE 2025, and 6th International Workshop, DeCaF 2025, Held in Conjunction with MICCAI 2025, Proceedings
A2 - Zamzmi, Ghada
A2 - Reinke, Annika
A2 - Samala, Ravi
A2 - Jiang, Meirui
A2 - Li, Xiaoxiao
A2 - Roth, Holger
A2 - Sidulova, Mariia
A2 - Kooi, Thijs
A2 - Albarqouni, Shadi
A2 - Bakas, Spyridon
A2 - Rieke, Nicola
PB - Springer Science and Business Media Deutschland GmbH
T2 - 1st International Workshop on Bridging Regulatory Science and Medical Imaging Evaluation, BRIDGE 2025 and 6th MICCAI Workshop on Distributed, Collaborative and Federated Learning, DeCaF 2025, Held in Conjunction with 28th International conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025
Y2 - 23 September 2025 through 27 September 2025
ER -