-
Je něco špatně v tomto záznamu ?
Explainable time-to-progression predictions in multiple sclerosis
R. D'hondt, K. Dedja, S. Aerts, B. Van Wijmeersch, T. Kalincik, S. Reddel, EK. Havrdova, A. Lugaresi, B. Weinstock-Guttman, S. Mrabet, P. Lalive, AG. Kermode, S. Ozakbas, F. Patti, A. Prat, V. Tomassini, I. Roos, R. Alroughani, O. Gerlach, SJ....
Jazyk angličtina Země Irsko
Typ dokumentu časopisecké články
- MeSH
- algoritmy MeSH
- časové faktory MeSH
- dospělí MeSH
- lidé MeSH
- prognóza MeSH
- progrese nemoci * MeSH
- registrace MeSH
- ROC křivka MeSH
- roztroušená skleróza * patofyziologie MeSH
- strojové učení * MeSH
- Check Tag
- dospělí MeSH
- lidé MeSH
- mužské pohlaví MeSH
- ženské pohlaví MeSH
- Publikační typ
- časopisecké články MeSH
BACKGROUND: Prognostic machine learning research in multiple sclerosis has been mainly focusing on black-box models predicting whether a patients' disability will progress in a fixed number of years. However, as this is a binary yes/no question, it cannot take individual disease severity into account. Therefore, in this work we propose to model the time to disease progression instead. Additionally, we use explainable machine learning techniques to make the model outputs more interpretable. METHODS: A preprocessed subset of 29,201 patients of the international data registry MSBase was used. Disability was assessed in terms of the Expanded Disability Status Scale (EDSS). We predict the time to significant and confirmed disability progression using random survival forests, a machine learning model for survival analysis. Performance is evaluated on a time-dependent area under the receiver operating characteristic and the precision-recall curves. Importantly, predictions are then explained using SHAP and Bellatrex, two explainability toolboxes, and lead to both global (population-wide) as well as local (patient visit-specific) insights. RESULTS: On the task of predicting progression in 2 years, the random survival forest achieves state-of-the-art performance, comparable to previous work employing a random forest. However, here the random survival forest has the added advantage of being able to predict progression over a longer time horizon, with AUROC >60% for the first 10 years after baseline. Explainability techniques further validated the model by extracting clinically valid insights from the predictions made by the model. For example, a clear decline in the per-visit probability of progression is observed in more recent years since 2012, likely reflecting globally increasing use of more effective MS therapies. CONCLUSION: The binary classification models found in the literature can be extended to a time-to-event setting without loss of performance, thus allowing a more comprehensive prediction of patient prognosis. Furthermore, explainability techniques proved to be key to reach a better understanding of the model and increase validation of its behaviour.
1 Biostat Data Science Institute Hasselt University Diepenbeek Belgium
Academic MS Center Zuyd Department of Neurology Zuyderland Medical Center Sittard Geleen Netherlands
Azienda Ospedaliera di Rilievo Nazionale San Giuseppe Moscati Avellino Avellino Italy
Biocruces Bizkaia Health Research Institute Spain
Bombay Hospital Institute of Medical Sciences Mumbai India
Centre for Molecular Medicine and Innovative Therapeutics Murdoch University Perth Australia
CHUM MS Center and Universite de Montreal Montreal Canada
College of Medicine and Health Sciences and Sultan Qaboos University Hospital Oman
CORe Department of Medicine University of Melbourne Melbourne Australia
CSSS Saint Jérôme Saint Jerome Canada
Department of Immunology Biomedical Research Institute Hasselt University Diepenbeek Belgium
Department of Medical and Surgical Sciences and Advanced Technologies GF Ingrassia Catania Italy
Department of Neurology Centro Hospitalar Universitario de Sao Joao Porto Portugal
Department of Neurology Cliniques Universitaires Saint Luc Brussels Belgium
Department of Neurology Concord Clinical School Concord Hospital Sydney Australia
Department of Neurology Concord Repatriation General Hospital Sydney Australia
Department of Neurology Faculty of Medicine University of Debrecen Debrecen Hungary
Department of Neurology Galliera Hospital Genova Italy
Department of Neurology Jacobs MS center for treatment and research United States
Department of Neurology Royal Brisbane and Women's Hospital Brisbane Australia
Department of Neurology The Alfred Hospital Melbourne Australia
Department of Neurology Universitary Hospital Ghent Ghent Belgium
Department of Neuroscience School of Translational Medicine Monash University Melbourne Australia
Department of Rehabilitation ML Novarese Hospital Moncrivello Moncrivello Italy
Dipartimento di Scienze Biomediche e Neuromotorie Università di Bologna Bologna Italy
Division of Neurology Department of Medicine Amiri Hospital Sharq Kuwait
Faculty of Medicine of Tunis University of Tunis El Manar Tunis Tunisia
FCS UFP Faculdade de Ciências da Saúde Portugal
FP I3ID Instituto de Investigação Inovação e Desenvolvimento Fernando Pessoa Portugal
Groene Hart Ziekenhuis Gouda Netherlands
IRCCS Istituto delle Scienze Neurologiche di Bologna Bologna Italy
itec imec research group at KU Leuven Kortrijk Belgium
Izmir University of Economics Medical Point Hospital Izmir Turkey
Jahn Ferenc Teaching Hospital Budapest Hungary
KU Leuven Dept Public Health and Primary Care Kortrijk Belgium
MS Centre Clinical Neurology SS Annunziata University Hospital Chieti Italy
Multiple Sclerosis Research Association Izmir Turkey
Multiple Sclerosis Unit AOU Policlinico G Rodolico San Marco University of Catania Italy
Neuroimmunology Centre Department of Neurology Royal Melbourne Hospital Melbourne Australia
Neurology department Hospital Fernandez Capital Federal Argentina
Neurology Department King Fahad Specialist Hospital Dammam Saudi Arabia
Noorderhart Hospitals Rehabilitation and MS Centre Pelt Belgium
RISE UFP rede de Investigação em Saúde Universidade Fernando Pessoa Porto Portugal
Sir Charles Gairdner Hospital Perth Australia
Sultan Qaboos University Al Khodh Oman
UHasselt Rehabilitation Research Center Faculty of Rehabilitation Sciences Diepenbeek Belgium
Université Catholique de Louvain Belgium
University MS Centre Hasselt University Hasselt Pelt Belgium
Citace poskytuje Crossref.org
- 000
- 00000naa a2200000 a 4500
- 001
- bmc25009295
- 003
- CZ-PrNML
- 005
- 20250429134544.0
- 007
- ta
- 008
- 250415e20250206ie f 000 0|eng||
- 009
- AR
- 024 7_
- $a 10.1016/j.cmpb.2025.108624 $2 doi
- 035 __
- $a (PubMed)39965473
- 040 __
- $a ABA008 $b cze $d ABA008 $e AACR2
- 041 0_
- $a eng
- 044 __
- $a ie
- 100 1_
- $a D'hondt, Robbe $u KU Leuven, Dept. Public Health and Primary Care, Kortrijk, Belgium; itec, imec research group at KU Leuven, Kortrijk, Belgium. Electronic address: robbe.dhondt@kuleuven.be
- 245 10
- $a Explainable time-to-progression predictions in multiple sclerosis / $c R. D'hondt, K. Dedja, S. Aerts, B. Van Wijmeersch, T. Kalincik, S. Reddel, EK. Havrdova, A. Lugaresi, B. Weinstock-Guttman, S. Mrabet, P. Lalive, AG. Kermode, S. Ozakbas, F. Patti, A. Prat, V. Tomassini, I. Roos, R. Alroughani, O. Gerlach, SJ. Khoury, V. van Pesch, MJ. Sá, J. Prevost, D. Spitaleri, P. McCombe, C. Solaro, A. van der Walt, H. Butzkueven, G. Laureys, JL. Sánchez-Menoyo, K. de Gans, A. Al-Asmi, N. Deri, T. Csepany, T. Al-Harbi, WM. Carroll, C. Rozsa, B. Singhal, TA. Hardy, S. Ramanathan, L. Peeters, C. Vens, MSBase Study Group
- 520 9_
- $a BACKGROUND: Prognostic machine learning research in multiple sclerosis has been mainly focusing on black-box models predicting whether a patients' disability will progress in a fixed number of years. However, as this is a binary yes/no question, it cannot take individual disease severity into account. Therefore, in this work we propose to model the time to disease progression instead. Additionally, we use explainable machine learning techniques to make the model outputs more interpretable. METHODS: A preprocessed subset of 29,201 patients of the international data registry MSBase was used. Disability was assessed in terms of the Expanded Disability Status Scale (EDSS). We predict the time to significant and confirmed disability progression using random survival forests, a machine learning model for survival analysis. Performance is evaluated on a time-dependent area under the receiver operating characteristic and the precision-recall curves. Importantly, predictions are then explained using SHAP and Bellatrex, two explainability toolboxes, and lead to both global (population-wide) as well as local (patient visit-specific) insights. RESULTS: On the task of predicting progression in 2 years, the random survival forest achieves state-of-the-art performance, comparable to previous work employing a random forest. However, here the random survival forest has the added advantage of being able to predict progression over a longer time horizon, with AUROC >60% for the first 10 years after baseline. Explainability techniques further validated the model by extracting clinically valid insights from the predictions made by the model. For example, a clear decline in the per-visit probability of progression is observed in more recent years since 2012, likely reflecting globally increasing use of more effective MS therapies. CONCLUSION: The binary classification models found in the literature can be extended to a time-to-event setting without loss of performance, thus allowing a more comprehensive prediction of patient prognosis. Furthermore, explainability techniques proved to be key to reach a better understanding of the model and increase validation of its behaviour.
- 650 _2
- $a lidé $7 D006801
- 650 12
- $a progrese nemoci $7 D018450
- 650 12
- $a roztroušená skleróza $x patofyziologie $7 D009103
- 650 12
- $a strojové učení $7 D000069550
- 650 _2
- $a prognóza $7 D011379
- 650 _2
- $a ROC křivka $7 D012372
- 650 _2
- $a ženské pohlaví $7 D005260
- 650 _2
- $a mužské pohlaví $7 D008297
- 650 _2
- $a dospělí $7 D000328
- 650 _2
- $a časové faktory $7 D013997
- 650 _2
- $a algoritmy $7 D000465
- 650 _2
- $a registrace $7 D012042
- 655 _2
- $a časopisecké články $7 D016428
- 700 1_
- $a Dedja, Klest $u KU Leuven, Dept. Public Health and Primary Care, Kortrijk, Belgium; itec, imec research group at KU Leuven, Kortrijk, Belgium
- 700 1_
- $a Aerts, Sofie $u University MS Centre (UMSC), Hasselt University, Hasselt-Pelt, Belgium; Department of Immunology, Biomedical Research Institute (BIOMED), Hasselt University, Diepenbeek, Belgium; Noorderhart Hospitals, Rehabilitation and MS Centre, Pelt, Belgium; UHasselt, Rehabilitation Research Center (REVAL), Faculty of Rehabilitation Sciences, Diepenbeek, Belgium
- 700 1_
- $a Van Wijmeersch, Bart $u University MS Centre (UMSC), Hasselt University, Hasselt-Pelt, Belgium; Department of Immunology, Biomedical Research Institute (BIOMED), Hasselt University, Diepenbeek, Belgium; Noorderhart Hospitals, Rehabilitation and MS Centre, Pelt, Belgium; UHasselt, Rehabilitation Research Center (REVAL), Faculty of Rehabilitation Sciences, Diepenbeek, Belgium
- 700 1_
- $a Kalincik, Tomas $u Neuroimmunology Centre, Department of Neurology, Royal Melbourne Hospital, Melbourne, Australia; CORe, Department of Medicine, University of Melbourne, Melbourne, Australia
- 700 1_
- $a Reddel, Stephen $u Department of Neurology, Concord Repatriation General Hospital, Sydney, Australia
- 700 1_
- $a Havrdova, Eva Kubala $u Department of Neurology and Center of Clinical Neuroscience, First Faculty of Medicine, Charles University in Prague and General University Hospital, Prague, Czech Republic
- 700 1_
- $a Lugaresi, Alessandra $u Dipartimento di Scienze Biomediche e Neuromotorie, Università di Bologna, Bologna, Italy; IRCCS Istituto delle Scienze Neurologiche di Bologna, Bologna, Italy
- 700 1_
- $a Weinstock-Guttman, Bianca $u Department of Neurology, Jacobs MS center for treatment and research, United States
- 700 1_
- $a Mrabet, Saloua $u Department of Neurology, LR 18SP03, Clinical Investigation Centre Neurosciences and Mental Health, Razi University Hospital, Tunis, Tunisia; Faculty of Medicine of Tunis, University of Tunis El Manar, Tunis, Tunisia
- 700 1_
- $a Lalive, Patrice $u Department of Clinical Neurosciences, Division of Neurology, Unit of Neuroimmunology, Geneva University Hospitals and Faculty of Medicine, Geneva, Switzerland
- 700 1_
- $a Kermode, Allan G $u Perron Institute for Neurological and Translational Science, The University of Western Australia, Perth, Australia; Centre for Molecular Medicine and Innovative Therapeutics, Murdoch University, Perth, Australia
- 700 1_
- $a Ozakbas, Serkan $u Izmir University of Economics, Medical Point Hospital, Izmir, Turkey; Multiple Sclerosis Research Association, Izmir, Turkey
- 700 1_
- $a Patti, Francesco $u Department of Medical and Surgical Sciences and Advanced Technologies, GF Ingrassia, Catania, Italy; Multiple Sclerosis Unit, AOU Policlinico "G Rodolico-San Marco", University of Catania, Italy
- 700 1_
- $a Prat, Alexandre $u CHUM MS Center and Universite de Montreal, Montreal, Canada
- 700 1_
- $a Tomassini, Valentina $u Institute for Advanced Biomedical Technologies (ITAB), Dept Neurosciences, Imaging and Clinical Sciences, University G. d'Annunzio of Chieti-Pescara, Chieti, Italy; MS Centre, Clinical Neurology, SS Annunziata University Hospital, Chieti, Italy
- 700 1_
- $a Roos, Izanne $u Neuroimmunology Centre, Department of Neurology, Royal Melbourne Hospital, Melbourne, Australia; CORe, Department of Medicine, University of Melbourne, Melbourne, Australia
- 700 1_
- $a Alroughani, Raed $u Division of Neurology, Department of Medicine, Amiri Hospital, Sharq, Kuwait
- 700 1_
- $a Gerlach, Oliver $u Academic MS Center Zuyd, Department of Neurology, Zuyderland Medical Center, Sittard-Geleen, Netherlands; School for Mental Health and Neuroscience, Department of Neurology, Maastricht University Medical Center, Maastricht 6131 BK, Netherlands
- 700 1_
- $a Khoury, Samia J $u Nehme and Therese Tohme Multiple Sclerosis Center, American University of Beirut Medical Centre, Beirut, Lebanon
- 700 1_
- $a van Pesch, Vincent $u Department of Neurology, Cliniques Universitaires Saint-Luc, Brussels, Belgium; Université Catholique de Louvain, Belgium
- 700 1_
- $a Sá, Maria José $u Department of Neurology, Centro Hospitalar Universitario de Sao Joao, Porto, Portugal; FP-I3ID, Instituto de Investigação, Inovação e Desenvolvimento Fernando Pessoa, Portugal; FCS-UFP, Faculdade de Ciências da Saúde, Portugal; RISE-UFP, rede de Investigação em Saúde, Universidade Fernando Pessoa, Porto, Portugal
- 700 1_
- $a Prevost, Julie $u CSSS Saint-Jérôme, Saint-Jerome, Canada
- 700 1_
- $a Spitaleri, Daniele $u Azienda Ospedaliera di Rilievo Nazionale San Giuseppe Moscati Avellino, Avellino, Italy
- 700 1_
- $a McCombe, Pamela $u Department of Neurology, Royal Brisbane and Women's Hospital, Brisbane, Australia; University of Queensland, Australia
- 700 1_
- $a Solaro, Claudio $u Department of Neurology, Galliera Hospital, Genova, Italy; Department of Rehabilitation, ML Novarese Hospital Moncrivello, Moncrivello, Italy
- 700 1_
- $a van der Walt, Anneke $u Department of Neurology, The Alfred Hospital, Melbourne, Australia; Department of Neuroscience, School of Translational Medicine, Monash University, Melbourne, Australia
- 700 1_
- $a Butzkueven, Helmut $u Department of Neurology, The Alfred Hospital, Melbourne, Australia; Department of Neuroscience, School of Translational Medicine, Monash University, Melbourne, Australia
- 700 1_
- $a Laureys, Guy $u Department of Neurology, Universitary Hospital Ghent, Ghent, Belgium
- 700 1_
- $a Sánchez-Menoyo, José Luis $u Department of Neurology, Galdakao-Usansolo University Hospital, Osakidetza-Basque Health Service, Galdakao, Spain; Biocruces-Bizkaia Health Research Institute, Spain
- 700 1_
- $a de Gans, Koen $u Groene Hart Ziekenhuis, Gouda, Netherlands
- 700 1_
- $a Al-Asmi, Abdullah $u Sultan Qaboos University, Al-Khodh, Oman; College of Medicine & Health Sciences and Sultan Qaboos University Hospital, Oman
- 700 1_
- $a Deri, Norma $u Neurology department, Hospital Fernandez, Capital Federal, Argentina
- 700 1_
- $a Csepany, Tunde $u Department of Neurology, Faculty of Medicine, University of Debrecen, Debrecen, Hungary
- 700 1_
- $a Al-Harbi, Talal $u Neurology Department, King Fahad Specialist Hospital-Dammam, Saudi Arabia
- 700 1_
- $a Carroll, William M $u Perron Institute for Neurological and Translational Science, The University of Western Australia, Perth, Australia; Sir Charles Gairdner Hospital, Perth, Australia
- 700 1_
- $a Rozsa, Csilla $u Jahn Ferenc Teaching Hospital, Budapest, Hungary
- 700 1_
- $a Singhal, Bhim $u Bombay Hospital Institute of Medical Sciences, Mumbai, India
- 700 1_
- $a Hardy, Todd A $u Department of Neurology, Concord Repatriation General Hospital, Sydney, Australia
- 700 1_
- $a Ramanathan, Sudarshini $u Translational Neuroimmunology Group, Kids Neuroscience Centre and Brain and Mind Centre, Faculty of Medicine and Health, University of Sydney, Sydney, Australia; Department of Neurology, Concord Clinical School, Concord Hospital, Sydney, Australia
- 700 1_
- $a Peeters, Liesbet $u University MS Centre (UMSC), Hasselt University, Hasselt-Pelt, Belgium; Department of Immunology, Biomedical Research Institute (BIOMED), Hasselt University, Diepenbeek, Belgium; I-Biostat, Data Science Institute (DSI), Hasselt University, Diepenbeek, Belgium
- 700 1_
- $a Vens, Celine $u KU Leuven, Dept. Public Health and Primary Care, Kortrijk, Belgium; itec, imec research group at KU Leuven, Kortrijk, Belgium
- 710 2_
- $a MSBase Study Group
- 773 0_
- $w MED00001214 $t Computer methods and programs in biomedicine $x 1872-7565 $g Roč. 263 (20250206), s. 108624
- 856 41
- $u https://pubmed.ncbi.nlm.nih.gov/39965473 $y Pubmed
- 910 __
- $a ABA008 $b sig $c sign $y - $z 0
- 990 __
- $a 20250415 $b ABA008
- 991 __
- $a 20250429134540 $b ABA008
- 999 __
- $a ok $b bmc $g 2310960 $s 1246376
- BAS __
- $a 3
- BAS __
- $a PreBMC-MEDLINE
- BMC __
- $a 2025 $b 263 $c - $d 108624 $e 20250206 $i 1872-7565 $m Computer methods and programs in biomedicine $n Comput Methods Programs Biomed $x MED00001214
- LZP __
- $a Pubmed-20250415
