{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"omics_type":["Unknown"],"volume":["8(1)"],"submitter":["Pirmani A"],"pubmed_abstract":["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 aga"],"journal":["NPJ digital medicine"],"pagination":["478"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12289948"],"repository":["biostudies-literature"],"pubmed_title":["Personalized federated learning for predicting disability progression in multiple sclerosis using real-world routine clinical data."],"pmcid":["PMC12289948"],"pubmed_authors":["Alroughani R","Singhal B","Kalincik T","Kuhle J","John N","Cauchi M","Guimaraes J","Garber J","De Brouwer E","Peeters LM","Willekens B","Rajda C","Hughes S","Tomassini V","Maimone D","Sas A","Lugaresi A","Taylor B","Van Wijmeersch B","Al-Asmi A","Sirbu CA","McCombe P","Castillo Trivino T","Havrdova EK","Passemiers A","Ozakbas S","de Gans K","Carroll WM","Pirmani A","Buzzard K","Hardy TA","Lalive P","Perez-Sempere A","Moreau Y","Ramanathan S","Van Pesch V","Rozsa C","Amato MP","Solaro C","Grammond P","Shuey N","Prat A","Boz C","Horakova D","Oldenhof M","Csepany T","Mason D","Blanco Y","Fabis-Pedrini M","Decoo D","Patti F","Cartechini E","Faes A","Arany A","Gouider R","Gerlach O","Lechner-Scott J","Simu M","Roos I","Laureys G","Altintas A","Spitaleri D","Kermode AG","Al-Harbi T","Prevost J","Gray O","Van der Walt A","Butzkueven H","Khoury SJ","Soysal A","Sanchez-Menoyo JL","Weinstock-Guttman B","Ampapa R"],"additional_accession":[]},"is_claimable":false,"name":"Personalized federated learning for predicting disability progression in multiple sclerosis using real-world routine clinical data.","description":"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 aga","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Jul","modification":"2026-03-27T17:04:07.93Z","creation":"2025-08-28T03:08:54.013Z"},"accession":"S-EPMC12289948","cross_references":{"pubmed":["40707601"],"doi":["10.1038/s41746-025-01788-8"]}}