Machine Learning Classification of Multiple Sclerosis in Children Using Optical Coherence Tomography

Overview

Journal Mult Scler

Publisher Sage Publications

Specialty Neurology

Date 2022 Aug 10

PMID 35946086

Authors

Beyza Ciftci Kavaklioglu

Lauren Erdman

Anna Goldenberg

Can Kavaklioglu

Cara Alexander

Hannah M Oppermann

Amish Patel

Soaad Hossain

Tara Berenbaum

Olivia Yau

Carmen Yea

Mina Ly

Fiona Costello

Jean K Mah

Arun Reginald

Brenda Banwell

Giulia Longoni

E Ann Yeh

Affiliations

Soon will be listed here.

Abstract

Background: In children, multiple sclerosis (MS) is the ultimate diagnosis in only 1/5 to 1/3 of cases after a first episode of central nervous system (CNS) demyelination. As the visual pathway is frequently affected in MS and other CNS demyelinating disorders (DDs), structural retinal imaging such as optical coherence tomography (OCT) can be used to differentiate MS.

Objective: This study aimed to investigate the utility of machine learning (ML) based on OCT features to identify distinct structural retinal features in children with DDs.

Methods: This study included 512 eyes from 187 ( = 374) children with demyelinating diseases and 69 ( = 138) controls. Input features of the analysis comprised of 24 auto-segmented OCT features.

Results: Random Forest classifier with recursive feature elimination yielded the highest predictive values and identified DDs with 75% and MS with 80% accuracy, while multiclass distinction between MS and monophasic DD was performed with 64% accuracy. A set of eight retinal features were identified as the most important features in this classification.

Conclusion: This study demonstrates that ML based on OCT features can be used to support a diagnosis of MS in children.

Citing Articles

New Method of Early RRMS Diagnosis Using OCT-Assessed Structural Retinal Data and Explainable Artificial Intelligence.

Ortiz M, Pueyo A, Dongil F, Boquete L, Sanchez-Morla E, Barea R Transl Vis Sci Technol. 2025; 14(2):14.

PMID: 39928305 PMC: 11812612. DOI: 10.1167/tvst.14.2.14.

Diagnosis of multiple sclerosis using optical coherence tomography supported by explainable artificial intelligence.

Dongil-Moreno F, Ortiz M, Pueyo A, Boquete L, Sanchez-Morla E, Jimeno-Huete D Eye (Lond). 2024; 38(8):1502-1508.

PMID: 38297153 PMC: 11126721. DOI: 10.1038/s41433-024-02933-5.

References

. Grading diabetic retinopathy from stereoscopic color fundus photographs--an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991; 98(5 Suppl):786-806. View

Waldman A, Hiremath G, Avery R, Conger A, Pineles S, Loguidice M . Monocular and binocular low-contrast visual acuity and optical coherence tomography in pediatric multiple sclerosis. Mult Scler Relat Disord. 2014; 3(3):326-334. PMC: 3964624. DOI: 10.1016/j.msard.2013.10.008. View

Saxena R, Bandyopadhyay G, Singh D, Singh S, Sharma P, Menon V . Evaluation of changes in retinal nerve fiber layer thickness and visual functions in cases of optic neuritis and multiple sclerosis. Indian J Ophthalmol. 2013; 61(10):562-6. PMC: 3853452. DOI: 10.4103/0301-4738.121071. View

Chua J, Bostan M, Li C, Sim Y, Bujor I, Wong D . A multi-regression approach to improve optical coherence tomography diagnostic accuracy in multiple sclerosis patients without previous optic neuritis. Neuroimage Clin. 2022; 34:103010. PMC: 9043868. DOI: 10.1016/j.nicl.2022.103010. View

Costello F, Pan Y, Yeh E, Hodge W, Burton J, Kardon R . The temporal evolution of structural and functional measures after acute optic neuritis. J Neurol Neurosurg Psychiatry. 2015; 86(12):1369-73. DOI: 10.1136/jnnp-2014-309704. View

Banwell B, Kennedy J, Sadovnick D, Arnold D, Magalhaes S, Wambera K . Incidence of acquired demyelination of the CNS in Canadian children. Neurology. 2009; 72(3):232-9. DOI: 10.1212/01.wnl.0000339482.84392.bd. View

Costello F, Hodge W, Pan Y, Freedman M, DeMeulemeester C . Differences in retinal nerve fiber layer atrophy between multiple sclerosis subtypes. J Neurol Sci. 2009; 281(1-2):74-9. DOI: 10.1016/j.jns.2009.02.354. View

Yeh E, Weinstock-Guttman B, Lincoff N, Reynolds J, Weinstock A, Madurai N . Retinal nerve fiber thickness in inflammatory demyelinating diseases of childhood onset. Mult Scler. 2009; 15(7):802-10. DOI: 10.1177/1352458509104586. View

Brandt A, Specovius S, Oberwahrenbrock T, Zimmermann H, Paul F, Costello F . Frequent retinal ganglion cell damage after acute optic neuritis. Mult Scler Relat Disord. 2018; 22:141-147. DOI: 10.1016/j.msard.2018.04.006. View

10.

Yau G, Lee J, Lau P, Tam V, Wong W, Yuen C . Longitudinal Changes in Retinal Nerve Fibre Layer Thickness after an Isolated Unilateral Retrobulbar Optic Neuritis: 1-Year Results. Neuroophthalmology. 2016; 39(1):22-25. PMC: 5123179. DOI: 10.3109/01658107.2014.984230. View

11.

Krupp L, Tardieu M, Amato M, Banwell B, Chitnis T, Dale R . International Pediatric Multiple Sclerosis Study Group criteria for pediatric multiple sclerosis and immune-mediated central nervous system demyelinating disorders: revisions to the 2007 definitions. Mult Scler. 2013; 19(10):1261-7. DOI: 10.1177/1352458513484547. View

12.

Al-Louzi O, Bhargava P, Newsome S, Balcer L, Frohman E, Crainiceanu C . Outer retinal changes following acute optic neuritis. Mult Scler. 2015; 22(3):362-72. PMC: 4724567. DOI: 10.1177/1352458515590646. View

13.

Yilmaz U, Gucuyener K, Erin D, Yazar Z, Gurkas E, Serdaroglu A . Reduced retinal nerve fiber layer thickness and macular volume in pediatric multiple sclerosis. J Child Neurol. 2012; 27(12):1517-23. DOI: 10.1177/0883073812447683. View

14.

Eyre M, Hameed A, Wright S, Brownlee W, Ciccarelli O, Bowman R . Retinal nerve fibre layer thinning is associated with worse visual outcome after optic neuritis in children with a relapsing demyelinating syndrome. Dev Med Child Neurol. 2018; 60(12):1244-1250. DOI: 10.1111/dmcn.13757. View

15.

Graves J, Chohan H, Cedars B, Arnow S, Yiu H, Waubant E . Sex differences and subclinical retinal injury in pediatric-onset MS. Mult Scler. 2016; 23(3):447-455. DOI: 10.1177/1352458516652497. View

16.

Thompson A, Banwell B, Barkhof F, Carroll W, Coetzee T, Comi G . Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol. 2017; 17(2):162-173. DOI: 10.1016/S1474-4422(17)30470-2. View

17.

Wingerchuk D, Banwell B, Bennett J, Cabre P, Carroll W, Chitnis T . International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology. 2015; 85(2):177-89. PMC: 4515040. DOI: 10.1212/WNL.0000000000001729. View

18.

Narayan R, McCreary M, Conger D, Wang C, Greenberg B . Unique characteristics of optical coherence tomography (OCT) results and visual acuity testing in myelin oligodendrocyte glycoprotein (MOG) antibody positive pediatric patients. Mult Scler Relat Disord. 2018; 28:86-90. DOI: 10.1016/j.msard.2018.11.026. View

19.

Schnurman Z, Frohman T, Beh S, Conger D, Conger A, Saidha S . Retinal architecture and mfERG: Optic nerve head component response characteristics in MS. Neurology. 2014; 82(21):1888-96. PMC: 4105253. DOI: 10.1212/WNL.0000000000000447. View

20.

Chen Q, Zhao G, Huang Y, Li Z, Sun X, Lu P . Clinical Characteristics of Pediatric Optic Neuritis With Myelin Oligodendrocyte Glycoprotein Seropositive: A Cohort Study. Pediatr Neurol. 2018; 83:42-49. DOI: 10.1016/j.pediatrneurol.2018.03.003. View