RT Journal Article
T1 Longitudinal machine learning modeling of MS patient trajectories improves predictions of disability progression: [Computer Methods and Programs in Biomedicine, Volume 208, (September 2021) 106180].
A1 De-Brouwer, Edward
A1 Becker, Thijs
A1 Moreau, Yves
A1 Havrdova, Eva Kubala
A1 Trojano, Maria
A1 Eichau, Sara
A1 Ozakbas, Serkan
A1 Onofrj, Marco
A1 Grammond, Pierre
A1 Kuhle, Jens
A1 Kappos, Ludwig
A1 Sola, Patrizia
A1 Cartechini, Elisabetta
A1 Lechner-Scott, Jeannette
A1 Alroughani, Raed
A1 Gerlach, Oliver
A1 Kalincik, Tomas
A1 Granella, Franco
A1 Grand'Maison, Francois
A1 Bergamaschi, Roberto
A1 Sa, Maria Jose
A1 Van-Wijmeersch, Bart
A1 Soysal, Aysun
A1 Sanchez-Menoyo, Jose Luis
A1 Solaro, Claudio
A1 Boz, Cavit
A1 Iuliano, Gerardo
A1 Buzzard, Katherine
A1 Aguera-Morales, Eduardo
A1 Terzi, Murat
A1 Trivio, Tamara Castillo
A1 Spitaleri, Daniele
A1 Van-Pesch, Vincent
A1 Shaygannejad, Vahid
A1 Moore, Fraser
A1 Oreja-Guevara, Celia
A1 Maimone, Davide
A1 Gouider, Riadh
A1 Csepany, Tunde
A1 Ramo-Tello, Cristina
A1 Peeters, Liesbet
K1 Disability progression
K1 Electronic health records
K1 Longitudinal data
K1 Multiple sclerosis
K1 Real-world data
K1 Recurrent neural networks
AB Background and objectives: Research in Multiple Sclerosis (MS) has recently focused on extracting knowledge from real-world clinical data sources. This type of data is more abundant than data produced during clinical trials and potentially more informative about real-world clinical practice. However, this comes at the cost of less curated and controlled data sets. In this work we aim to predict disability progression by optimally extracting information from longitudinal patient data in the real-world setting, with a special focus on the sporadic sampling problem. Methods: We use machine learning methods suited for patient trajectories modeling, such as recurrent neural networks and tensor factorization. A subset of 6682 patients from the MSBase registry is used. Results: We can predict disability progression of patients in a two-year horizon with an ROC-AUC of 0.85, which represents a 32% decrease in the ranking pair error (1-AUC) compared to reference methods using static clinical features. Conclusions: Compared to the models available in the literature, this work uses the most complete patient history for MS disease progression prediction and represents a step forward towards AI-assisted precision medicine in MS.
PB Elsevier
YR 2021
FD 2021-06-17
LK http://hdl.handle.net/10668/22137
UL http://hdl.handle.net/10668/22137
LA en
NO De Brouwer E, Becker T, Moreau Y, Havrdova EK, Trojano M, Eichau S, et al. Longitudinal machine learning modeling of MS patient trajectories improves predictions of disability progression. Comput Methods Programs Biomed. 2021 Sep;208:106180
DS RISalud
RD Sep 13, 2025