Machine learning classifier to identify clinical and radiological features relevant to disability progression in multiple sclerosis.
Disability progression
Machine learning
Magnetic resonance imaging
Multiple sclerosis
Journal
Journal of neurology
ISSN: 1432-1459
Titre abrégé: J Neurol
Pays: Germany
ID NLM: 0423161
Informations de publication
Date de publication:
Dec 2021
Dec 2021
Historique:
received:
16
12
2020
accepted:
05
05
2021
revised:
05
05
2021
pubmed:
11
5
2021
medline:
5
11
2021
entrez:
10
5
2021
Statut:
ppublish
Résumé
To evaluate the accuracy of a data-driven approach, such as machine learning classification, in predicting disability progression in MS. We analyzed structural brain images of 163 subjects diagnosed with MS acquired at two different sites. Participants were followed up for 2-6 years, with disability progression defined according to the expanded disability status scale (EDSS) increment at follow-up. T2-weighted lesion load (T2LL), thalamic and cerebellar gray matter (GM) volumes, fractional anisotropy of the normal appearing white matter were calculated at baseline and included in supervised machine learning classifiers. Age, sex, phenotype, EDSS at baseline, therapy and time to follow-up period were also included. Classes were labeled as stable or progressed disability. Participants were randomly chosen from both sites to build a sample including 50% patients showing disability progression and 50% patients being stable. One-thousand machine learning classifiers were applied to the resulting sample, and after testing for overfitting, classifier confusion matrix, relative metrics and feature importance were evaluated. At follow-up, 36% of participants showed disability progression. The classifier with the highest resulting metrics had accuracy of 0.79, area under the true positive versus false positive rates curve of 0.81, sensitivity of 0.90 and specificity of 0.71. T2LL, thalamic volume, disability at baseline and administered therapy were identified as important features in predicting disability progression. Classifiers built on radiological features had higher accuracy than those built on clinical features. Disability progression in MS may be predicted via machine learning classifiers, mostly evaluating neuroradiological features.
Identifiants
pubmed: 33970338
doi: 10.1007/s00415-021-10605-7
pii: 10.1007/s00415-021-10605-7
pmc: PMC8563671
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
4834-4845Subventions
Organisme : Fondazione Italiana Sclerosi Multipla
ID : FISM/2018/S/3
Informations de copyright
© 2021. The Author(s).
Références
Confavreux C, Vukusic S (2014) The clinical course of multiple sclerosis. Handb Clin Neurol 122:343–369. https://doi.org/10.1016/B978-0-444-52001-2.00014-5
doi: 10.1016/B978-0-444-52001-2.00014-5
pubmed: 24507525
Ciccarelli O, Barkhof F, Bodini B et al (2014) Pathogenesis of multiple sclerosis: insights from molecular and metabolic imaging. Lancet Neurol 13:807–822. https://doi.org/10.1016/S1474-4422(14)70101-2
doi: 10.1016/S1474-4422(14)70101-2
pubmed: 25008549
Amato MP, Fonderico M, Portaccio E et al (2020) Disease-modifying drugs can reduce disability progression in relapsing multiple sclerosis. Brain. https://doi.org/10.1093/brain/awaa251
doi: 10.1093/brain/awaa251
pubmed: 32935843
Tommasin S, Giannì C, De Giglio L, Pantano P (2017) Neuroimaging techniques to assess inflammation in multiple sclerosis. Neuroscience. https://doi.org/10.1016/j.neuroscience.2017.07.055
doi: 10.1016/j.neuroscience.2017.07.055
pubmed: 28764938
Tintore M, Rovira À, Río J et al (2015) Defining high, medium and low impact prognostic factors for developing multiple sclerosis. Brain 138:1863–1874. https://doi.org/10.1093/brain/awv105
doi: 10.1093/brain/awv105
pubmed: 25902415
Pontillo G, Cocozza S, Di Stasi M et al (2020) 2D linear measures of ventricular enlargement may be relevant markers of brain atrophy and long-term disability progression in multiple sclerosis. Eur Radiol. https://doi.org/10.1007/s00330-020-06738-4
doi: 10.1007/s00330-020-06738-4
pubmed: 32100089
Radue E-W, Barkhof F, Kappos L et al (2015) Correlation between brain volume loss and clinical and MRI outcomes in multiple sclerosis. Neurology 84:784–793. https://doi.org/10.1212/WNL.0000000000001281
doi: 10.1212/WNL.0000000000001281
pubmed: 25632085
pmcid: 4339126
Louapre C, Bodini B, Lubetzki C et al (2017) Imaging markers of multiple sclerosis prognosis. Curr Opin Neurol 30:231–236. https://doi.org/10.1097/WCO.0000000000000456
doi: 10.1097/WCO.0000000000000456
pubmed: 28362719
Filippi M, Brück W, Chard D et al (2019) Association between pathological and MRI findings in multiple sclerosis. Lancet Neurol 18:198–210. https://doi.org/10.1016/S1474-4422(18)30451-4
doi: 10.1016/S1474-4422(18)30451-4
pubmed: 30663609
Eshaghi A, Prados F, Brownlee WJ et al (2018) Deep gray matter volume loss drives disability worsening in multiple sclerosis. Ann Neurol 83:210–222. https://doi.org/10.1002/ana.25145
doi: 10.1002/ana.25145
pubmed: 29331092
pmcid: 5838522
Cocozza S, Petracca M, Mormina E et al (2017) Cerebellar lobule atrophy and disability in progressive MS. J Neurol Neurosurg Psychiatry 88:1065–1072. https://doi.org/10.1136/jnnp-2017-316448
doi: 10.1136/jnnp-2017-316448
pubmed: 28844067
D’Ambrosio A, Pagani E, Riccitelli GC et al (2017) Cerebellar contribution to motor and cognitive performance in multiple sclerosis: an MRI sub-regional volumetric analysis. Mult Scler 23:1194–1203. https://doi.org/10.1177/1352458516674567
doi: 10.1177/1352458516674567
pubmed: 27760859
Patti F, De Stefano M, Lavorgna L et al (2015) Lesion load may predict long-term cognitive dysfunction in multiple sclerosis patients. PLoS ONE 10:e0120754. https://doi.org/10.1371/journal.pone.0120754
doi: 10.1371/journal.pone.0120754
pubmed: 25816303
pmcid: 4376682
Calabrese M, Poretto V, Favaretto A et al (2012) Cortical lesion load associates with progression of disability in multiple sclerosis. Brain 135:2952–2961. https://doi.org/10.1093/brain/aws246
doi: 10.1093/brain/aws246
pubmed: 23065788
Wilkins A (2017) Cerebellar dysfunction in multiple sclerosis. Front Neurol. https://doi.org/10.3389/fneur.2017.00312
doi: 10.3389/fneur.2017.00312
pubmed: 29375465
pmcid: 5487391
Rocca MA, Mesaros S, Pagani E et al (2010) Thalamic damage and long-term progression of disability in multiple sclerosis. Radiology 257:463–469. https://doi.org/10.1148/radiol.10100326
doi: 10.1148/radiol.10100326
pubmed: 20724544
Datta G, Colasanti A, Rabiner EA et al (2017) Neuroinflammation and its relationship to changes in brain volume and white matter lesions in multiple sclerosis. Brain 140:2927–2938. https://doi.org/10.1093/brain/awx228
doi: 10.1093/brain/awx228
pubmed: 29053775
Wottschel V, Alexander DC, Kwok PP et al (2015) Predicting outcome in clinically isolated syndrome using machine learning. Neuroimage Clin 7:281–287. https://doi.org/10.1016/j.nicl.2014.11.021
doi: 10.1016/j.nicl.2014.11.021
pubmed: 25610791
Zurita M, Montalba C, Labbé T et al (2018) Characterization of relapsing-remitting multiple sclerosis patients using support vector machine classifications of functional and diffusion MRI data. Neuroimage Clin 20:724–730. https://doi.org/10.1016/j.nicl.2018.09.002
doi: 10.1016/j.nicl.2018.09.002
pubmed: 30238916
pmcid: 6148733
Polman CH, Reingold SC, Banwell B et al (2011) Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 69:292–302. https://doi.org/10.1002/ana.22366
doi: 10.1002/ana.22366
pubmed: 21387374
pmcid: 3084507
Thompson AJ, Banwell BL, Barkhof F et al (2018) Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 17:162–173. https://doi.org/10.1016/S1474-4422(17)30470-2
doi: 10.1016/S1474-4422(17)30470-2
pubmed: 29275977
Río J, Nos C, Tintoré M et al (2006) Defining the response to interferon-beta in relapsing-remitting multiple sclerosis patients. Ann Neurol 59:344–352. https://doi.org/10.1002/ana.20740
doi: 10.1002/ana.20740
pubmed: 16437558
Raz E, Cercignani M, Sbardella E et al (2009) Clinically isolated syndrome suggestive of multiple sclerosis: voxelwise regional investigation of White and Gray matter. Radiology 254:227–234. https://doi.org/10.1148/radiol.2541090817
doi: 10.1148/radiol.2541090817
pubmed: 20019140
Zhao Y, Healy BC, Rotstein D et al (2017) Exploration of machine learning techniques in predicting multiple sclerosis disease course. PLoS ONE 12:e0174866. https://doi.org/10.1371/journal.pone.0174866
doi: 10.1371/journal.pone.0174866
pubmed: 28379999
pmcid: 5381810
Bluemke DA, Moy L, Bredella MA et al (2020) Assessing radiology research on artificial intelligence: a brief guide for authors, reviewers, and readers—from the radiology editorial board. Radiology 294:487–489. https://doi.org/10.1148/radiol.2019192515
doi: 10.1148/radiol.2019192515
pubmed: 31891322
Law MT, Traboulsee AL, Li DK et al (2019) Machine learning in secondary progressive multiple sclerosis: an improved predictive model for short-term disability progression. Mult Scler J Exp Transl Clin 5:2055217319885983. https://doi.org/10.1177/2055217319885983
doi: 10.1177/2055217319885983
pubmed: 31723436
pmcid: 6836306
Gasperini C, Prosperini L, Tintoré M et al (2019) Unraveling treatment response in multiple sclerosis: a clinical and MRI challenge. Neurology 92:180–192. https://doi.org/10.1212/WNL.0000000000006810
doi: 10.1212/WNL.0000000000006810
pubmed: 30587516
pmcid: 6345120
Šimundić A-M (2009) Measures of diagnostic accuracy: basic definitions. EJIFCC 19:203–211
pubmed: 27683318
pmcid: 4975285
Schoonheim MM, Geurts JJG, Barkhof F (2010) The limits of functional reorganization in multiple sclerosis. Neurology 74:1246–1247. https://doi.org/10.1212/WNL.0b013e3181db9957
doi: 10.1212/WNL.0b013e3181db9957
pubmed: 20404304
Calabrese M, Mattisi I, Rinaldi F et al (2010) Magnetic resonance evidence of cerebellar cortical pathology in multiple sclerosis. J Neurol Neurosurg Psychiatry 81:401–404. https://doi.org/10.1136/jnnp.2009.177733
doi: 10.1136/jnnp.2009.177733
pubmed: 19965849
Davie CA, Barker GJ, Webb S et al (1995) Persistent functional deficit in multiple sclerosis and autosomal dominant cerebellar ataxia is associated with axon loss. Brain 118(Pt 6):1583–1592. https://doi.org/10.1093/brain/118.6.1583
doi: 10.1093/brain/118.6.1583
pubmed: 8595487
Barkhof F (2002) The clinico-radiological paradox in multiple sclerosis revisited. Curr Opin Neurol 15:239–245
doi: 10.1097/00019052-200206000-00003
Elliott C, Belachew S, Wolinsky JS et al (2019) Chronic white matter lesion activity predicts clinical progression in primary progressive multiple sclerosis. Brain 142:2787–2799. https://doi.org/10.1093/brain/awz212
doi: 10.1093/brain/awz212
pubmed: 31497864
pmcid: 6736181
Tintore M, Arrambide G, Otero-Romero S et al (2019) The long-term outcomes of CIS patients in the Barcelona inception cohort: Looking back to recognize aggressive MS. Mult Scler. https://doi.org/10.1177/1352458519877810
doi: 10.1177/1352458519877810
pubmed: 31674875
pmcid: 7604549
Bakshi R, Healy BC, Dupuy SL et al (2020) Brain MRI predicts worsening multiple sclerosis disability over 5 years in the SUMMIT study. J Neuroimaging. https://doi.org/10.1111/jon.12688
doi: 10.1111/jon.12688
pubmed: 33351983
pmcid: 7194808
Dwyer M, Brior D, Lyman C et al (2020) Artificial intelligence-based thalamic volumetry is fast, reliable, and generalizable to large, heterogeneous datasets using only clinical quality T2-FLAIR MRI (4846). Neurology 94:4846
Azevedo CJ, Cen SY, Khadka S et al (2018) Thalamic atrophy in multiple sclerosis: a magnetic resonance imaging marker of neurodegeneration throughout disease. Ann Neurol 83:223–234. https://doi.org/10.1002/ana.25150
doi: 10.1002/ana.25150
pubmed: 29328531
pmcid: 6317847
Tona F, Petsas N, Sbardella E et al (2014) Multiple sclerosis: altered thalamic resting-state functional connectivity and its effect on cognitive function. Radiology 271:814–821. https://doi.org/10.1148/radiol.14131688
doi: 10.1148/radiol.14131688
pubmed: 24484065
Stankoff B, Louapre C (2018) Can we use regional grey matter atrophy sequence to stage neurodegeneration in multiple sclerosis? Brain 141:1580–1583. https://doi.org/10.1093/brain/awy114
doi: 10.1093/brain/awy114
pubmed: 29800475
Haines JD, Inglese M, Casaccia P (2011) axonal damage in multiple sclerosis. Mt Sinai J Med 78:231–243. https://doi.org/10.1002/msj.20246
doi: 10.1002/msj.20246
pubmed: 21425267
pmcid: 3142952
Kolasa M, Hakulinen U, Brander A et al (2019) Diffusion tensor imaging and disability progression in multiple sclerosis: a 4-year follow-up study. Brain Behav 9:e01194. https://doi.org/10.1002/brb3.1194
doi: 10.1002/brb3.1194
pubmed: 30588771
Kalincik T, Manouchehrinia A, Sobisek L et al (2017) Towards personalized therapy for multiple sclerosis: prediction of individual treatment response. Brain 140:2426–2443. https://doi.org/10.1093/brain/awx185
doi: 10.1093/brain/awx185
pubmed: 29050389
Cree BAC, Mares J, Hartung H-P (2019) Current therapeutic landscape in multiple sclerosis: an evolving treatment paradigm. Curr Opin Neurol 32:365–377. https://doi.org/10.1097/WCO.0000000000000700
doi: 10.1097/WCO.0000000000000700
pubmed: 30985372
Hart FM, Bainbridge J (2016) Current and emerging treatment of multiple sclerosis. Am J Manag Care 22:s159-170
pubmed: 27356025
Cohen JA, Reingold SC, Polman CH et al (2012) Disability outcome measures in multiple sclerosis clinical trials: current status and future prospects. Lancet Neurol 11:467–476. https://doi.org/10.1016/S1474-4422(12)70059-5
doi: 10.1016/S1474-4422(12)70059-5
pubmed: 22516081