Artificial intelligence (AI) is the branch of science aiming at creating algorithms able to carry out tasks that typically require human intelligence. In medicine, there has been a tremendous increase in AI applications thanks to increasingly powerful computers and the emergence of big data repositories. Multiple sclerosis (MS) is a chronic autoimmune condition affecting the central nervous system with a complex pathogenesis, a challenging diagnostic process strongly relying on magnetic resonance imaging (MRI) and a high and largely unexplained variability across patients. Therefore, AI applications in MS have the great potential of helping us better support the diagnosis, find markers for prognosis to eventually design more powerful randomised clinical trials and improve patient management in clinical practice and eventually understand the mechanisms of the disease. This topical review aims to summarise the recent advances in AI applied to MRI data in MS to illustrate its achievements, limitations and future directions.
McCullochWSPittsW. A logical calculus of the ideas immanent in nervous activity. Bull Math Biol1990; 52(1–2): 99–115; discussion 73–97.
2.
MorrisRGM. D.O. Hebb: the organization of behavior, Wiley: New York; 1949. Brain Res Bull1999; 50: 437.
3.
TuringAM. Computing machinery and intelligence. Mind1950; 49: 433–460.
4.
TurCColloroneS. Artificial intelligence in multiple sclerosis. Kranion2023; 18: 139–144.
5.
BonacchiRFilippiMRoccaMA. Role of artificial intelligence in MS clinical practice. NeuroImage Clin2022; 35: 103065.
6.
NajiYMahdaouiMKlevorR, et al. Artificial intelligence and multiple sclerosis: Up-to-date review. Cureus2023; 15(9): e45412.
7.
ThompsonAJBanwellBLBarkhofF, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol2018; 17: 162–173.
8.
EshaghiAWottschelVCorteseR, et al. Gray matter MRI differentiates neuromyelitis optica from multiple sclerosis using random forest. Neurology2016; 87: 2463–2470.
9.
RoccaMAAnzaloneNStorelliL, et al. Deep learning on conventional magnetic resonance imaging improves the diagnosis of multiple sclerosis mimics. Invest Radiol2021; 56: 252–260.
10.
ZuritaMMontalbaCLabbéT, et al. Characterization of relapsing-remitting multiple sclerosis patients using support vector machine classifications of functional and diffusion MRI data. NeuroImage Clin2018; 20: 724–730.
11.
EitelFSoehlerEBellmann-StroblJ, et al. Uncovering convolutional neural network decisions for diagnosing multiple sclerosis on conventional MRI using layer-wise relevance propagation. NeuroImage Clin2019; 24: 102003.
12.
Mato-AbadVLabiano-FontcubertaARodríguez-YáñezS, et al. Classification of radiologically isolated syndrome and clinically isolated syndrome with machine-learning techniques. Eur J Neurol2019; 26(7): 1000–1005.
13.
SeokJMChoWChungYH, et al. Differentiation between multiple sclerosis and neuromyelitis optica spectrum disorder using a deep learning model. Sci Rep2023; 13: 11625.
14.
TintoreMRoviraÀRíoJ, et al. Defining high, medium and low impact prognostic factors for developing multiple sclerosis. Brain2015; 138(Pt 7): 1863–1874.
15.
ChungKKAltmannDBarkhofF, et al. A 30-year clinical and magnetic resonance imaging observational study of multiple sclerosis and clinically isolated syndromes. Ann Neurol2020; 87(1): 63–74.
16.
ScalfariARomualdiCNicholasRS, et al. The cortical damage, early relapses, and onset of the progressive phase in multiple sclerosis. Neurology2018; 90: e2107–e2118.
17.
BrownleeWJAltmannDRPradosF, et al. Early imaging predictors of long-term outcomes in relapse-onset multiple sclerosis. Brain2019; 142: 2276–2287.
18.
WottschelVChardDTEnzingerC, et al. SVM recursive feature elimination analyses of structural brain MRI predicts near-term relapses in patients with clinically isolated syndromes suggestive of multiple sclerosis. NeuroImage Clin2019; 24: 102011.
19.
BendfeldtKTaschlerBGaetanoL, et al. MRI-based prediction of conversion from clinically isolated syndrome to clinically definite multiple sclerosis using SVM and lesion geometry. Brain Imaging Behav2019; 13(5): 1361–1374.
20.
ParetoDGarcia-VidalAGroppaS, et al. Prognosis of a second clinical event from baseline MRI in patients with a CIS: A multicenter study using a machine learning approach. Neuroradiology2022; 64(7): 1383–1390.
21.
TousignantALemaîtrePPrecupD, et al. Prediction of disease progression in multiple sclerosis patients using deep learning analysis of MRI data. Proc Mach Learn Res2019; 102: 483–492.
22.
RocaPAttyeAColasL, et al. Artificial intelligence to predict clinical disability in patients with multiple sclerosis using FLAIR MRI. Diagn Interv Imaging2020; 101(12): 795–802.
23.
StorelliLAzzimontiMGueyeM, et al. A deep learning approach to predicting disease progression in multiple sclerosis using magnetic resonance imaging. Invest Radiol2022; 57: 423–432.
24.
ZhaoYHealyBCRotsteinD, et al. Exploration of machine learning techniques in predicting multiple sclerosis disease course. PLoS ONE2017; 12(4): e0174866.
25.
HaiderLChungKBirchG, et al. Linear brain atrophy measures in multiple sclerosis and clinically isolated syndromes: A 30-year follow-up. J Neurol Neurosurg Psychiatry2021; 92: 839–846.
26.
CommowickOIstaceAKainM, et al. Objective evaluation of multiple sclerosis lesion segmentation using a data management and processing infrastructure. Sci Rep2018; 8: 13650.
27.
RoviraACorralJFAugerC, et al. Assessment of automatic decision-support systems for detecting active T2 lesions in multiple sclerosis patients. Mult Scler2022; 28(8): 1209–1218.
28.
Diaz-HurtadoMMartínez-HerasESolanaE, et al. Recent advances in the longitudinal segmentation of multiple sclerosis lesions on magnetic resonance imaging: A review. Neuroradiology2022; 64(11): 2103–2117.
29.
ZengCGuLLiuZ, et al. Review of deep learning approaches for the segmentation of multiple sclerosis lesions on brain MRI. Front Neuroinform2020; 14: 610967.
30.
SalemMValverdeSCabezasM, et al. A fully convolutional neural network for new T2-w lesion detection in multiple sclerosis. NeuroImage Clin2020; 25: 102149.
31.
SalemMRyanMAOliverA, et al. Improving the detection of new lesions in multiple sclerosis with a cascaded 3D fully convolutional neural network approach. Front Neurosci2022; 16: 1007619.
32.
NarayanaPACoronadoISujitSJ, et al. Deep learning for predicting enhancing lesions in multiple sclerosis from noncontrast MRI. Radiology2020; 294(2): 398–404.
33.
CabaBCafaroALombardA, et al. Single-timepoint low-dimensional characterization and classification of acute versus chronic multiple sclerosis lesions using machine learning. NeuroImage2023; 265: 119787.
34.
KhajetashBTalebiABagherpourZ, et al. Introducing radiomics model to predict active plaque in multiple sclerosis patients using magnetic resonance images. Biomed Phys Eng Express2023; 9: 055004.
35.
TavakoliHPirzad JahromiGSedaghatA. Investigating the ability of radiomics features for diagnosis of the active plaque of multiple sclerosis patients. J Biomed Phys Eng2023; 13(5): 421–432.
36.
ShekariFVardAAdibiI, et al. Investigating the feasibility of differentiating MS active lesions from inactive ones using texture analysis and machine learning methods in DWI images. Mult Scler Relat Disord2024; 82: 105363.
37.
AlijamaatANikravanShalmaniABayatP. Multiple sclerosis lesion segmentation from brain MRI using U-Net based on wavelet pooling. Int J Comput Assist Radiol Surg2021; 16(9): 1459–1467.
38.
DanelakisATheoharisTVerganelakisDA. Survey of automated multiple sclerosis lesion segmentation techniques on magnetic resonance imaging. Comput Med Imaging Graph2018; 70: 83–100.
39.
FartariaMJSatiPTodeaA, et al. Automated detection and segmentation of multiple sclerosis lesions using ultra–high-field MP2RAGE. Invest Radiol2019; 54(6): 356–364.
40.
FilippiMBrückWChardD, et al. Association between pathological and MRI findings in multiple sclerosis. Lancet Neurol2019; 18: 198–210.
41.
BarkhofF. The clinico-radiological paradox in multiple sclerosis revisited. Curr Opin Neurol2002; 15(3): 239–245.
42.
EshaghiAYoungALWijeratnePA, et al. Identifying multiple sclerosis subtypes using unsupervised machine learning and MRI data. Nat Commun2021; 12: 2078.
43.
YoungALMarinescuRVOxtobyNP, et al. Uncovering the heterogeneity and temporal complexity of neurodegenerative diseases with subtype and stage inference. Nat Commun2018; 9: 4273.
44.
PontilloGPennaSCocozzaS, et al. Stratification of multiple sclerosis patients using unsupervised machine learning: A single-visit MRI-driven approach. Eur Radiol2022; 32(8): 5382–5391.
45.
CollLParetoDCarbonell-MirabentP, et al. Deciphering multiple sclerosis disability with deep learning attention maps on clinical MRI. NeuroImage Clin2023; 38: 103376.
46.
YaminMAValsasinaPTessadoriJ, et al. Discovering functional connectivity features characterizing multiple sclerosis phenotypes using explainable artificial intelligence. Hum Brain Mapp2023; 44: 2294–2306.
47.
CorteseRColloroneSCiccarelliO, et al. Advances in brain imaging in multiple sclerosis. Ther Adv Neurol Disord2019; 12: 1756286419859722.
48.
ColloroneSPradosFKanberB, et al. Brain microstructural and metabolic alterations detected in vivo at onset of the first demyelinating event. Brain2021; 144: 1409–1421.
49.
CastelvecchiD. Can we open the black box of AI?Nature2016; 538: 20–23.
50.
CollinsGSDhimanPAndaur NavarroCL, et al. Protocol for development of a reporting guideline (TRIPOD-AI) and risk of bias tool (PROBAST-AI) for diagnostic and prognostic prediction model studies based on artificial intelligence. BMJ Open2021; 11: e048008.
51.
FaletJPRDurso-FinleyJNichyporukB, et al. Estimating individual treatment effect on disability progression in multiple sclerosis using deep learning. Nat Commun2022; 13: 5645.
52.
MuehlematterUJBluethgenCVokingerKN. FDA-cleared artificial intelligence and machine learning-based medical devices and their 510(k) predicate networks. Lancet Digit Health2023; 5(9): e618–e626.
53.
van LeeuwenKGSchalekampSRuttenMJCM, et al. Artificial intelligence in radiology: 100 commercially available products and their scientific evidence. Eur Radiol2021; 31(6): 3797–3804.
