Machine-learning-based prediction of disability progression in multiple sclerosis: An observational, international, multi-center study.
Journal: PLOS digital health
Volume: 3
Issue: 7
Year of Publication:
Affiliated Institutions:
ESAT-STADIUS, KU Leuven, Belgium.
I-Biostat, Hasselt University, Belgium.
SUMO, IDLAB, Ghent University - imec, Belgium.
KERMIT, Department of Data Analysis and Mathematical Modelling, Ghent University, Belgium.
Department of Neurology, Ghent University, Belgium.
Noorderhart ziekenhuizen Pelt, Belgium.
Charles University in Prague and General University Hospital, Prague, Czech Republic.
Department of Medical and Surgical Sciences and Advanced Technologies, GF Ingrassia, Catania, Italy.
Hospital Universitario Virgen Macarena, Sevilla, Spain.
CHUM and Université de Montreal, Montreal, Canada.
IRCCS Istituto delle Scienze Neurologiche di Bologna, Bologna, Italia and Dipartimento di Scienze Biomediche e Neuromotorie, Università di Bologna, Bologna, Italia.
CISSS Chaudière-Appalache, Levis, Canada.
Melbourne MS Centre, Department of Neurology, Royal Melbourne Hospital, Melbourne, Australia.
Amiri Hospital, Sharq, Kuwait.
Neuro Rive-Sud, Quebec, Canada.
Box Hill Hospital, Melbourne, Australia.
Mayis University, Samsun, Turkey.
University Newcastle, Newcastle, Australia.
Academic MS Center Zuyderland, Department of Neurology, Zuyderland Medical Center, Sittard-Geleen, The Netherlands.
American University of Beirut Medical Center, Beirut, Lebanon.
Azienda Sanitaria Unica Regionale Marche - AV, Macerata, Italy.
Cliniques Universitaires Saint-Luc, Brussels, Belgium.
Centro Hospitalar Universitario de Sao Joao, Porto, Portugal.
Department of Neurology, Buffalo General Medical Center, Buffalo, United States of America.
Hospital Clinic de Barcelona, Barcelona, Spain.
Nemocnice Jihlava, Jihlava, Czech Republic.
Azienda Ospedaliera di Rilievo Nazionale San Giuseppe Moscati Avellino, Avellino, Italy.
Dept. of Rehabilitation, CRFF Mons. Luigi Novarese, Moncrivello, Italy.
MS center, UOC Neurologia, ARNAS Garibaldi, Catania, Italy.
Bakirkoy Education and Research Hospital for Psychiatric and Neurological Diseases, Istanbul, Turkey.
Ospedali Riuniti di Salerno, Salerno, Italy.
Razi Hospital, Manouba, Tunisia.
Hospital Universitario Donostia, San Sebastián, Spain.
Hospital de Galdakao-Usansolo, Galdakao, Spain.
Universitary Hospital Ghent, Ghent, Belgium.
The Alfred Hospital, Melbourne, Australia.
St. Michael's Hospital, Toronto, Canada.
University Hospital Reina Sofia, Cordoba, Spain.
Koc University, School of Medicine, Istanbul, Turkey.
College of Medicine & Health Sciences and Sultan Qaboos University Hospital, SQU, Oman.
Groene Hart Ziekenhuis, Gouda, Netherlands.
Universidade Metropolitana de Santos, Santos, Brazil.
University of Debrecen, Debrecen, Hungary.
Liverpool Hospital, Sydney, Australia.
Hospital Fernandez, Capital Federal, Argentina.
King Fahad Specialist Hospital-Dammam, Khobar, Saudi Arabia.
Royal Hobart Hospital, Hobart, Australia.
South Eastern HSC Trust, Belfast, United Kingdom.
Geneva University Hospital, Geneva, Switzerland.
Jahn Ferenc Teaching Hospital, Budapest, Hungary.
St Vincent's University Hospital, Dublin, Ireland.
University of Western Australia, Nedlands, Australia.
Hospital General Universitario de Alicante, Alicante, Spain.
Emergency Clinical County Hospital Pius Brinzeu, Timisoara, Romania and University of Medicine and Pharmacy Victor Babes, Timisoara, Romania.
Semmelweis University Budapest, Budapest, Hungary.
Concord Repatriation General Hospital, Sydney, Australia.
AZ Alma Ziekenhuis, Sijsele - Damme, Belgium.
Royal Victoria Hospital, Belfast, United Kingdom.
AHEPA University Hospital, Thessaloniki, Greece.
BAZ County Hospital, Miskolc, Hungary.
Mater Dei Hospital, Msida, Malta.
Data Science Institute, Hasselt University, Belgium.
Abstract summary
Disability progression is a key milestone in the disease evolution of people with multiple sclerosis (PwMS). Prediction models of the probability of disability progression have not yet reached the level of trust needed to be adopted in the clinic. A common benchmark to assess model development in multiple sclerosis is also currently lacking.Data of adult PwMS with a follow-up of at least three years from 146 MS centers, spread over 40 countries and collected by the MSBase consortium was used. With basic inclusion criteria for quality requirements, it represents a total of 15, 240 PwMS. External validation was performed and repeated five times to assess the significance of the results. Transparent Reporting for Individual Prognosis Or Diagnosis (TRIPOD) guidelines were followed. Confirmed disability progression after two years was predicted, with a confirmation window of six months. Only routinely collected variables were used such as the expanded disability status scale, treatment, relapse information, and MS course. To learn the probability of disability progression, state-of-the-art machine learning models were investigated. The discrimination performance of the models is evaluated with the area under the receiver operator curve (ROC-AUC) and under the precision recall curve (AUC-PR), and their calibration via the Brier score and the expected calibration error. All our preprocessing and model code are available at https://gitlab.com/edebrouwer/ms_benchmark, making this task an ideal benchmark for predicting disability progression in MS.Machine learning models achieved a ROC-AUC of 0⋅71 ± 0⋅01, an AUC-PR of 0⋅26 ± 0⋅02, a Brier score of 0⋅1 ± 0⋅01 and an expected calibration error of 0⋅07 ± 0⋅04. The history of disability progression was identified as being more predictive for future disability progression than the treatment or relapses history.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.
Authors & Co-authors:
De Brouwer Edward E
Becker Thijs T
Werthen-Brabants Lorin L
Dewulf Pieter P
Iliadis Dimitrios D
Dekeyser Cathérine C
Laureys Guy G
Van Wijmeersch Bart B
Popescu Veronica V
Dhaene Tom T
Deschrijver Dirk D
Waegeman Willem W
De Baets Bernard B
Stock Michiel M
Horakova Dana D
Patti Francesco F
Izquierdo Guillermo G
Eichau Sara S
Girard Marc M
Prat Alexandre A
Lugaresi Alessandra A
Grammond Pierre P
Kalincik Tomas T
Alroughani Raed R
Grand'Maison Francois F
Skibina Olga O
Terzi Murat M
Lechner-Scott Jeannette J
Gerlach Oliver O
Khoury Samia J SJ
Cartechini Elisabetta E
Van Pesch Vincent V
Sà Maria José MJ
Weinstock-Guttman Bianca B
Blanco Yolanda Y
Ampapa Radek R
Spitaleri Daniele D
Solaro Claudio C
Maimone Davide D
Soysal Aysun A
Iuliano Gerardo G
Gouider Riadh R
Castillo-Triviño Tamara T
Sánchez-Menoyo José Luis JL
Laureys Guy G
van der Walt Anneke A
Oh Jiwon J
Aguera-Morales Eduardo E
Altintas Ayse A
Al-Asmi Abdullah A
de Gans Koen K
Fragoso Yara Y
Csepany Tunde T
Hodgkinson Suzanne S
Deri Norma N
Al-Harbi Talal T
Taylor Bruce B
Gray Orla O
Lalive Patrice P
Rozsa Csilla C
McGuigan Chris C
Kermode Allan A
Sempere Angel Pérez AP
Mihaela Simu S
Simo Magdolna M
Hardy Todd T
Decoo Danny D
Hughes Stella S
Grigoriadis Nikolaos N
Sas Attila A
Vella Norbert N
Moreau Yves Y
Peeters Liesbet L
Study Outcome
Source Link: Visit source