Machine-learning-based prediction of disability progression in multiple sclerosis: An observational, international, multi-center study.
Edward De BrouwerThijs BeckerLorin Werthen-BrabantsPieter DewulfDimitrios IliadisCathérine DekeyserGuy LaureysBart Van WijmeerschVeronica PopescuTom DhaeneDirk DeschrijverWillem WaegemanBernard De BaetsMichiel StockDana HorakovaFrancesco PattiGuillermo IzquierdoSara EichauMarc GirardAlexandre PratAlessandra LugaresiPierre GrammondTomas KalincikRaed AlroughaniFrancois Grand'MaisonOlga SkibinaMurat TerziJeannette Lechner-ScottOliver GerlachSamia J KhouryElisabetta CartechiniVincent Van PeschMaria José SàBianca Weinstock-GuttmanYolanda BlancoRadek AmpapaDaniele SpitaleriClaudio SolaroDavide MaimoneAysun SoysalGerardo IulianoRiadh GouiderTamara Castillo TriviñoJosé Luis Sánchez-MenoyoGuy LaureysAnneke van der WaltJiwon OhEduardo Aguera-MoralesAyse AltintasAbdullah Al-AsmiKoen de GansYara FragosoTünde CsépánySuzanne HodgkinsonNorma DeriTalal Al-HarbiBruce TaylorOrla GrayPatrice LaliveCsilla RozsaChris McGuiganAllan G KermodeAngel Pérez SempereSimu MihaelaMagdolna SimoTodd HardyDanny DecooStella HughesNikolaos GrigoriadisAttila SasNorbert VellaYves MoreauLiesbet M PeetersPublished in: PLOS digital health (2024)
Good discrimination and calibration performance on an external validation set is achieved, using only routinely collected variables. This suggests machine-learning models can reliably inform clinicians about the future occurrence of progression and are mature for a clinical impact study.