Deep Learning on Conventional Magnetic Resonance Imaging Improves the Diagnosis of Multiple Sclerosis Mimics.
Journal
Investigative radiology
ISSN: 1536-0210
Titre abrégé: Invest Radiol
Pays: United States
ID NLM: 0045377
Informations de publication
Date de publication:
01 04 2021
01 04 2021
Historique:
pubmed:
29
10
2020
medline:
16
10
2021
entrez:
28
10
2020
Statut:
ppublish
Résumé
The aims of this study were to present a deep learning approach for the automated classification of multiple sclerosis and its mimics and compare model performance with that of 2 expert neuroradiologists. A total of 268 T2-weighted and T1-weighted brain magnetic resonance imagin scans were retrospectively collected from patients with migraine (n = 56), multiple sclerosis (n = 70), neuromyelitis optica spectrum disorders (n = 91), and central nervous system vasculitis (n = 51). The neural network architecture, trained on 178 scans, was based on a cascade of 4 three-dimensional convolutional layers, followed by a fully dense layer after feature extraction. The ability of the final algorithm to correctly classify the diseases in an independent test set of 90 scans was compared with that of the neuroradiologists. The interrater agreement was 84.9% (Cohen κ = 0.78, P < 0.001). In the test set, deep learning and expert raters reached the highest diagnostic accuracy in multiple sclerosis (98.8% vs 72.8%, P < 0.001, for rater 1; and 81.8%, P < 0.001, for rater 2) and the lowest in neuromyelitis optica spectrum disorders (88.6% vs 4.4%, P < 0.001, for both raters), whereas they achieved intermediate values for migraine (92.2% vs 53%, P = 0.03, for rater 1; and 64.8%, P = 0.01, for rater 2) and vasculitis (92.1% vs 54.6%, P = 0.3, for rater 1; and 45.5%, P = 0.2, for rater 2). The overall performance of the automated method exceeded that of expert raters, with the worst misdiagnosis when discriminating between neuromyelitis optica spectrum disorders and vasculitis or migraine. A neural network performed better than expert raters in terms of accuracy in classifying white matter disorders from magnetic resonance imaging and may help in their diagnostic work-up.
Identifiants
pubmed: 33109920
pii: 00004424-202104000-00007
doi: 10.1097/RLI.0000000000000735
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
252-260Informations de copyright
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.
Déclaration de conflit d'intérêts
Conflicts of interest and sources of funding: none declared.
Références
Thompson AJ, Banwell BL, Barkhof F, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol . 2018;17:162–173.
van der Vuurst de Vries RM, Mescheriakova JY, Wong YYM, et al. Application of the 2017 revised McDonald criteria for multiple sclerosis to patients with a typical clinically isolated syndrome. JAMA Neurol . 2018;75:1392–1398.
Hyun JW, Kim W, Huh SY, et al. Application of the 2017 McDonald diagnostic criteria for multiple sclerosis in Korean patients with clinically isolated syndrome. Mult Scler . 2019;25:1488–1495.
Kaisey M, Solomon AJ, Luu M, et al. Incidence of multiple sclerosis misdiagnosis in referrals to two academic centers. Mult Scler Relat Disord . 2019;30:51–56.
Solomon AJ, Bourdette DN, Cross AH, et al. The contemporary spectrum of multiple sclerosis misdiagnosis: a multicenter study. Neurology . 2016;87:1393–1399.
Kim SS, Richman DP, Johnson WO, et al. Limited utility of current MRI criteria for distinguishing multiple sclerosis from common mimickers: primary and secondary CNS vasculitis, lupus and Sjogren's syndrome. Mult Scler . 2014;20:57–63.
Absinta M, Rocca MA, Colombo B, et al. Patients with migraine do not have MRI-visible cortical lesions. J Neurol . 2012;259:2695–2698.
Liu S, Kullnat J, Bourdette D, et al. Prevalence of brain magnetic resonance imaging meeting Barkhof and McDonald criteria for dissemination in space among headache patients. Mult Scler . 2013;19:1101–1105.
Cacciaguerra L, Meani A, Mesaros S, et al. Brain and cord imaging features in neuromyelitis optica spectrum disorders. Ann Neurol . 2019;85:371–384.
Solomon AJ, Naismith RT, Cross AH. Misdiagnosis of multiple sclerosis: impact of the 2017 McDonald criteria on clinical practice. Neurology . 2019;92:26–33.
Jurynczyk M, Tackley G, Kong Y, et al. Brain lesion distribution criteria distinguish MS from AQP4-antibody NMOSD and MOG-antibody disease. J Neurol Neurosurg Psychiatry . 2017;88:132–136.
Matthews L, Marasco R, Jenkinson M, et al. Distinction of seropositive NMO spectrum disorder and MS brain lesion distribution. Neurology . 2013;80:1330–1337.
Maggi P, Absinta M, Grammatico M, et al. Central vein sign differentiates multiple sclerosis from central nervous system inflammatory vasculopathies. Ann Neurol . 2018;83:283–294.
Lakhani P, Prater AB, Hutson RK, et al. Machine learning in radiology: applications beyond image interpretation. J Am Coll Radiol . 2018;15:350–359.
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature . 2015;521:436–444.
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal . 2017;42:60–88.
Jo T, Nho K, Saykin AJ. Deep learning in Alzheimer's disease: diagnostic classification and prognostic prediction using neuroimaging data. Front Aging Neurosci . 2019;11:220.
Vieira S, Pinaya WHL, Mechelli A. Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: methods and applications. Neurosci Biobehav Rev . 2017;74(pt A):58–75.
Eitel F, Soehler E, Bellmann-Strobl J, et al. Uncovering convolutional neural network decisions for diagnosing multiple sclerosis on conventional MRI using layer-wise relevance propagation. Neuroimage Clin . 2019;24:102003.
Zhang YWL, Wei G. Multiple sclerosis identification by convolutional neural network with dropout and parametric ReLU. J Comput Sci . 2018;28:1–10.
Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders: 2nd edition. Cephalalgia . 2004;24(suppl 1):9–160.
Wingerchuk DM, Banwell B, Bennett JL, et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology . 2015;85:177–189.
Kim HJ, Paul F, Lana-Peixoto MA, et al. MRI characteristics of neuromyelitis optica spectrum disorder: an international update. Neurology . 2015;84:1165–1173.
Filippi M, Preziosa P, Banwell BL, et al. Assessment of lesions on magnetic resonance imaging in multiple sclerosis: practical guidelines. Brain . 2019;142:1858–1875.
Karen Simonyan AZ. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556. 2014.
Kosinski AS. A weighted generalized score statistic for comparison of predictive values of diagnostic tests. Stat Med . 2013;32:964–977.
Mafee MF, Rapoport M, Karimi A, et al. Orbital and ocular imaging using 3- and 1.5-T MR imaging systems. Neuroimaging Clin N Am . 2005;15:1–21.
Calabrese M, Oh MS, Favaretto A, et al. No MRI evidence of cortical lesions in neuromyelitis optica. Neurology . 2012;79:1671–1676.
Filippi M, Rocca MA, Calabrese M, et al. Intracortical lesions: relevance for new MRI diagnostic criteria for multiple sclerosis. Neurology . 2010;75:1988–1994.
Beck ES, Gai N, Filippini S, et al. Inversion recovery susceptibility weighted imaging with enhanced T2 weighting at 3 T improves visualization of subpial cortical multiple sclerosis lesions [published ahead of print June 22, 2020]. Invest Radiol .
Hagiwara A, Kamagata K, Aoki S. Image domain transfer by deep learning is feasible in multiple sclerosis clinical practice. Invest Radiol . 2020;55:324–325.
Eichinger P, Hock A, Schon S, et al. Acceleration of double inversion recovery sequences in multiple sclerosis with compressed sensing. Invest Radiol . 2019;54:319–324.
Finck T, Li H, Grundl L, et al. Deep-learning generated synthetic double inversion recovery images improve multiple sclerosis lesion detection. Invest Radiol . 2020;55:318–323.
Porter A, Gladstone JP, Dodick DW. Migraine and white matter hyperintensities. Curr Pain Headache Rep . 2005;9:289–293.
Abdel Razek AAK, Alvarez H, Bagg S, et al. Imaging spectrum of CNS vasculitis. Radiographics . 2014;34:873–894.
Mader S, Jeganathan V, Arinuma Y, et al. Understanding the antibody repertoire in neuropsychiatric systemic lupus erythematosus and neuromyelitis optica spectrum disorder: do they share common targets? Arthritis Rheumatol . 2018;70:277–286.
Asgari N, Jarius S, Laustrup H, et al. Aquaporin-4-autoimmunity in patients with systemic lupus erythematosus: a predominantly population-based study. Mult Scler . 2018;24:331–339.
Estiasari R, Matsushita T, Masaki K, et al. Comparison of clinical, immunological and neuroimaging features between anti-aquaporin-4 antibody-positive and antibody-negative Sjogren's syndrome patients with central nervous system manifestations. Mult Scler . 2012;18:807–816.
Rauschecker AM, Rudie JD, Xie L, et al. Artificial intelligence system approaching neuroradiologist-level differential diagnosis accuracy at brain MRI. Radiology . 2020;295:626–637.
Bach S, Binder A, Montavon G, et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS One . 2015;10:e0130140.