Early prediction of disability progression in multiple sclerosis (MS) remains challenging despite its critical importance for therapeutic decision-making. We present the first systematic evaluation of personalized federated learning (PFL) for 2-year MS disability progression prediction, leveraging multi-center real-world data from over 26,000 patients. While conventional federated learning (FL) enables privacy-aware collaborative modeling, it remains vulnerable to institutional data heterogeneity. PFL overcomes this challenge by adapting shared models to local data distributions without compromising privacy. We evaluated two personalization strategies: a novel AdaptiveDualBranchNet architecture with selective parameter sharing, and personalized fine-tuning of global models, benchmarked against centralized and client-specific approaches. Baseline FL underperformed relative to personalized methods, whereas personalization significantly improved performance, with personalized FedProx and FedAVG achieving ROC-AUC scores of 0.8398 ± 0.0019 and 0.8384 ± 0.0014, respectively. These findings establish personalization as critical for scalable, privacy-aware clinical prediction models and highlight its potential to inform earlier intervention strategies in MS and beyond.

Academic MS Center Zuyd Department of Neurology Zuyderland Medical Center Sittard Geleen The Netherlands

AZ Alma Ziekenhuis Damme Belgium

Azienda Ospedaliera di Rilievo Nazionale San Giuseppe Moscati Avellino Avellino Italy

Bakirkoy Education and Research Hospital for Psychiatric and Neurological Diseases Istanbul Turkey

Biomedical Research Institute Hasselt University Hasselt Belgium

Bombay Hospital Institute of Medical Sciences Mumbai India

Centro Sclerosi Multipla UOC Neurologia Azienda Opsedaliera per l'Emergenza Cannizzaro Catania Italy

Christchurch Hospital Christchurch New Zealand

CHUM and Universite de Montreal Montreal QC Canada

CISSS Chaudière Appalache Levis QC Canada

Clinical Neurosciences Department 'Carol Davila' University of Medicine and Pharmacy Bucharest Romania

College of Medicine and Health Sciences Sultan Qaboos University Al Khodh Oman

CSSS Saint Jérôme Saint Jerome QC Canada

Data Science Institute Hasselt University Hasselt Belgium

Department NEUROFARBA University of Florence Florence Italy

Department of Clinical Neurosciences and Mental Health Faculty of Medicine of University of Porto Porto Portugal

Department of Clinical Neurosciences Division of Neurology Unit of Neuroimmunology Geneva University Hospitals and Faculty of Medicine Geneva Switzerland

Department of Medical and Surgical Sciences and Advanced Technologies GF Ingrassia Catania Italy

Department of Medicine School of Clinical Sciences Monash University Clayton VIC Australia

Department of Neurology and Center of Clinical Neuroscience 1st Faculty of Medicine Charles University Prague and General University Hospital Prague Czechia

Department of Neurology and Stroke BAZ County Hospital Miskolc Hungary

Department of Neurology Antwerp University Hospital Edegem Belgium

Department of Neurology Cliniques Universitaires Saint Luc Brussels Belgium

Department of Neurology Concord Repatriation General Hospital Sydney NSW Australia

Department of Neurology Galdakao Usansolo University Hospital Osakidetza Basque Health Service Galdakao Spain

Department of Neurology Jacobs MS Center for Treatment and Research New York NY USA

Department of Neurology LR 18SP03 Clinical Investigation Centre Neurosciences and Mental Health Razi University Hospital Tunis Tunisia

Department of Neurology Medical Faculty Karadeniz Technical University Trabzon Turkey

Department of Neurology Neuroimmunology Centre Royal Melbourne Hospital Melbourne VIC Australia

Department of Neurology Royal Brisbane Hospital Brisbane QLD Australia

Department of Neurology School of Medicine and Koc University Research Center for Translational Medicine Koc University Istanbul Turkey

Department of Neurology The Alfred Hospital Melbourne VIC Australia

Department of Neurology Unidade Local de Saúde de São João Porto Portugal

Department of Neurology Universitary Hospital Ghent Ghent Belgium

Department of Neurology University Hospital and University of Basel Basel Switzerland

Department of Neurology University of Szeged Szeged Hungary

Department of Neurology Westmead Hospital Sydney NSW Australia

Department of Neurosciences Box Hill Hospital Box Hill VIC Australia

Dipartimento di Scienze Biomediche e Neuromotorie Università di Bologna Bologna Italy

Division of Neurology Department of Medicine Amiri Hospital Sharq Kuwait

Faculty of Medicine University of Debrecen Debrecen Hungary

Groene Hart Ziekenhuis Gouda The Netherlands

Hospital Universitario Donostia and IIS Biodonostia San Sebastián Spain

Hunter Medical Research Institute University Newcastle Newcastle NSW Australia

Institute for Advanced Biomedical Technologies Dept Neurosciences Imaging and Clinical Sciences University G d'Annunzio of Chieti Pescara Chieti Italy

IRCCS Fondazione Don Carlo Gnocchi Florence Italy

Jahn Ferenc Teaching Hospital Budapest Hungary

Medical Point Hospital Izmir University of Economics Izmir Turkey

Nehme and Therese Tohme Multiple Sclerosis Center American University of Beirut Medical Center Beirut Lebanon

Nemocnice Jihlava Jihlava Czechia

Neurology Department King Fahad Specialist Hospital Dammam Dammam Saudi Arabia

Neurology Unit AST Macerata Macerata Italy

Neurology Unit Galliera Hospital Genova Italy

Neurology Unit Hospital General Universitario de Alicante Alicante Spain

Neurosciences Department Mater Dei Hospital Birkirkara Malta

Noorderhart Hospitals Rehabilitation and MS University MS Centre Hasselt Pelt Belgium

Perron Institute for Neurological and Translational Science Sir Charles Gairdner Hospital The University of Western Australia Perth WA Australia

Perron Institute for Neurological and Translational Science The University of Western Australia Perth WA Australia

Perron Institute QEII Medical Centre University of Western Australia Nedlands WA Australia

Royal Hobart Hospital Hobart TAS Australia

Royal Victoria Hospital Belfast UK

Service of Neurology Center of Neuroimmunology Hospital Clinic de Barcelona Barcelona Spain

South Eastern HSC Trust Belfast UK

St Vincents Hospital Fitzroy Melbourne VIC Australia

STADIUS ESAT KU Leuven Leuven Belgium

Translational Neuroimmunology Group Kids Neuroscience Centre and Brain and Mind Centre Faculty of Medicine and Health University of Sydney Sydney NSW Australia

University Multiple Sclerosis Center Hasselt University Hasselt Belgium

University of Antwerp Antwerp Belgium

University of Medicine and Pharmacy Victor Babes Timisoara Timisoara Romania

See more in PubMed

Walton, C. et al. Rising prevalence of multiple sclerosis worldwide: insights from the atlas of MS, third edition. PubMed PMC

McGinley, M. P., Goldschmidt, C. H. & Rae-Grant, A. D. Diagnosis and treatment of multiple sclerosis: a review. PubMed

Reich, D. S., Lucchinetti, C. F. & Calabresi, P. A. Multiple sclerosis. PubMed PMC

Degenhardt, A., Ramagopalan, S. V., Scalfari, A. & Ebers, G. C. Clinical prognostic factors in multiple sclerosis: a natural history review. PubMed

Pellegrini, F. et al. Predicting disability progression in multiple sclerosis: insights from advanced statistical modeling. PubMed

Seker, B. I. O. et al. Prognostic models for predicting clinical disease progression, worsening and activity in people with multiple sclerosis. PubMed PMC

Sherman, R. E. et al. Real-world evidence—what is it and what can it tell us? PubMed

Brown, F. S. et al. Systematic review of prediction models in relapsing remitting multiple sclerosis. PubMed PMC

Havas, J. et al. Predictive medicine in multiple sclerosis: a systematic review. PubMed

Seccia, R. et al. Machine learning use for prognostic purposes in multiple sclerosis. PubMed PMC

Hartmann, M., Fenton, N. & Dobson, R. Current review and next steps for artificial intelligence in multiple sclerosis risk research. PubMed

Brouwer, E. D. et al. Longitudinal modeling of MS patient trajectories improves predictions of disability progression. PubMed

De Brouwer, E. et al. Machine-learning-based prediction of disability progression in multiple sclerosis: an observational, international, multi-center study. PubMed PMC

Pirmani, A. et al. The journey of data within a global data sharing initiative: a federated 3-layer data analysis pipeline to scale up multiple sclerosis research. PubMed PMC

Jensen, P. B., Jensen, L. J. & Brunak, S. Mining electronic health records: towards better research applications and clinical care. PubMed

Wu, J., Roy, J. & Stewart, W. F. Prediction modeling using EHR data: challenges, strategies, and a comparison of machine learning approaches. PubMed

Weiskopf, N. G. & Weng, C. Methods and dimensions of electronic health record data quality assessment: enabling reuse for clinical research. PubMed PMC

Wilkinson, M. D. et al. The fair guiding principles for scientific data management and stewardship. PubMed PMC

Antunes, R. S., André da Costa, C., Küderle, A., Yari, I. A. & Eskofier, B. Federated learning for healthcare: systematic review and architecture proposal.

Xu, J. et al. Federated learning for healthcare informatics. PubMed PMC

Li, S. et al. Federated and distributed learning applications for electronic health records and structured medical data: a scoping review. PubMed PMC

Rieke, N. et al. The future of digital health with federated learning. PubMed PMC

Brisimi, T. S. et al. Federated learning of predictive models from federated electronic health records. PubMed PMC

Yin, X., Zhu, Y. & Hu, J. A comprehensive survey of privacy-preserving federated learning: a taxonomy, review, and future directions.

Wang, W. et al. A privacy preserving framework for federated learning in smart healthcare systems.

Truex, S. et al. A hybrid approach to privacy-preserving federated learning.

Donkada, S. et al. Uncovering promises and challenges of federated learning to detect cardiovascular diseases: a scoping literature review (2023).

Yi, L. et al. Su-net: an efficient encoder-decoder model of federated learning for brain tumor segmentation. In

Oldenhof, M. et al. Industry-scale orchestrated federated learning for drug discovery.

Liu, D. et al. Multiple sclerosis lesion segmentation: revisiting weighting mechanisms for federated learning. PubMed PMC

Denissen, S. et al. Towards multimodal machine learning prediction of individual cognitive evolution in multiple sclerosis. PubMed PMC

Denissen, S. et al. Transfer learning on structural brain age models to decode cognition in MS: a federated learning approach.

Tan, A. Z., Yu, H., Cui, L. & Yang, Q. Towards personalized federated learning. PubMed

Butzkueven, H. et al. Msbase: an international, online registry and platform for collaborative outcomes research in multiple sclerosis. PubMed

Kalincik, T. & Butzkueven, H. The MSBase registry: informing clinical practice. PubMed

McMahan, H. B., Moore, E., Ramage, D., Hampson, S. & y Arcas, B. A. Communication-efficient learning of deep networks from decentralized data https://arxiv.org/abs/1602.05629 (2023).

Li, T., Sanjabi, M., Zaheer, M., Talwalkar, A. & Smith, V. On the convergence of federated optimization in heterogeneous networks (2018).

Reddi, S. J. et al. Adaptive federated optimization. In

Biewald, L. Experiment tracking with weights and biases https://www.wandb.com/. Software available from wandb.com (2020).

Guo, C., Pleiss, G., Sun, Y. & Weinberger, K. Q. On calibration of modern neural networks. In:

Davis, S. E., Greevy, R. A., Lasko, T. A., Walsh, C. G. & Matheny, M. E. Detection of calibration drift in clinical prediction models to inform model updating. PubMed PMC

Arivazhagan, M. G., Aggarwal, V., Singh, A. K. & Choudhary, S. Federated learning with personalization layers https://arxiv.org/abs/1912.00818 (2019).

PyTorch. ReduceLROnPlateau—PyTorch 2.6 documentation. https://pytorch.org/docs/stable/generated/torch.optim.lr_scheduler.ReduceLROnPlateau.html. Accessed 28 Mar 2025 (2019).

Khodak, M. et al. Federated hyperparameter tuning: challenges, baselines, and connections to weight-sharing https://arxiv.org/abs/2106.04502 (2021).

Zawad, S. & Yan, F. Hyperparameter tuning for federated learning—systems and practices. In Nguyen, L. M., Hoang, T. N. & Chen, P.-Y. (eds.)

Zhang, H. et al. Federated learning hyperparameter tuning from a system perspective.

Yosinski, J., Clune, J., Bengio, Y. & Lipson, H. How transferable are features in deep neural networks? In:

Houlsby, N. et al. Parameter-efficient transfer learning for NLP. DOI

Howard, J. & Ruder, S. Universal language model fine-tuning for text classification. In

Kairouz, P. et al. Advances and open problems in federated learning.

Li, T., Sahu, A. K., Talwalkar, A. & Smith, V. Federated learning: challenges, methods, and future directions.

Pirmani, A., Moreau, Y. & Peeters, L. M. Unlocking the power of real-world data: a framework for sustainable healthcare. PubMed

EU. European Health Data Space regulation (EHDS). https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space-regulation-ehds_en. Accessed: 26 Mar 2025.

Auffray, C. et al. Making sense of big data in health research: Towards an EU action plan. PubMed PMC

Collins, G. S., Reitsma, J. B., Altman, D. G. & Moons, K. G. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (tripod): the tripod statement. PubMed PMC

Peeters, L. M. et al. Covid-19 in people with multiple sclerosis: a global data sharing initiative. PubMed PMC

Heyndrickx, W. et al. Melloddy: cross-pharma federated learning at unprecedented scale unlocks benefits in QSAR without compromising proprietary information. PubMed PMC

Andorra, M. et al. Predicting disease severity in multiple sclerosis using multimodal data and machine learning. PubMed PMC

Dwork, C. Differential privacy. In Bugliesi, M., Preneel, B., Sassone, V. & Wegener, I. (eds.)

Goldreich, O. Secure multi-party computation.

López-Blanco, R., Alonso, R. S., González-Arrieta, A., Chamoso, P. & Prieto, J. Federated learning of explainable artificial intelligence (FED-XAI): a review. In Ossowski, S. et al. (eds.)

Choi, A. et al. A novel deep learning algorithm for real-time prediction of clinical deterioration in the emergency department for a multimodal clinical decision support system. PubMed PMC

Kalincik, T. et al. Towards personalized therapy for multiple sclerosis: prediction of individual treatment response. PubMed

Leray, E. et al. Evidence for a two-stage disability progression in multiple sclerosis. PubMed PMC

Andersson, P. B., Waubant, E., Gee, L. & Goodkin, D. E. Multiple sclerosis that is progressive from the time of onset: clinical characteristics and progression of disability. PubMed

Liu, M., Qi, Y., Wang, W. & Sun, X. Toward a better understanding about real-world evidence. PubMed PMC

Pirmani, A., Oldenhof, M., Peeters, L. M., De Brouwer, E. & Moreau, Y. Accessible ecosystem for clinical research (federated learning for everyone): development and usability study. PubMed PMC

Paszke, A. et al. Pytorch: an imperative style, high-performance deep learning library. In:

Beutel, D. J. et al. Flower: a friendly federated learning research framework 10.48550/arXiv.2007.14390 (2020).