Machine learning classification of multiple sclerosis in children using optical coherence tomography

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

doi:10.1177/13524585221112605

Machine learning classification of multiple sclerosis in children using optical coherence tomography

Mult Scler. 2022 Dec;28(14):2253-2262. doi: 10.1177/13524585221112605. Epub 2022 Aug 9.

Authors

Beyza Ciftci Kavaklioglu¹, Lauren Erdman², Anna Goldenberg³, Can Kavaklioglu⁴, Cara Alexander⁵, Hannah M Oppermann⁶, Amish Patel⁵, Soaad Hossain⁷, Tara Berenbaum⁸, Olivia Yau⁸, Carmen Yea⁸, Mina Ly⁸, Fiona Costello⁹, Jean K Mah¹⁰, Arun Reginald¹¹, Brenda Banwell¹², Giulia Longoni¹³, E Ann Yeh^{13

14}

Affiliations

¹ Neuroscience and Mental Health Program, SickKids Research Institute, The Hospital for Sick Children, Toronto, ON, Canada/Department of Internal Medicine, Max Rady College of Medicine, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB, Canada.
² Department of Computer Science, University of Toronto, Toronto, ON, Canada; Vector Institute, Toronto, ON, Canada.
³ Department of Computer Science, University of Toronto, Toronto, ON, Canada; Vector Institute, Toronto, ON, Canada/Temerty Centre for AI Research and Education in Medicine, University of Toronto, Toronto, ON, Canada.
⁴ Department of Mechanical and Industrial Engineering, Ryerson University, Toronto, ON, Canada.
⁵ Department of Computer Science, University of Toronto, Toronto, ON, Canada.
⁶ Department of Computer Science, University of Toronto, Toronto, ON, Canada/Department of Information and Computing Sciences, Utrecht University, Utrecht, the Netherlands.
⁷ Department of Computer Science, University of Toronto, Toronto, ON, Canada/Temerty Centre for AI Research and Education in Medicine, University of Toronto, Toronto, ON, Canada/Environics Analytics, Toronto, ON, Canada.
⁸ Division of Neurology, Department of Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, ON, Canada.
⁹ Department of Clinical Neurosciences, Hotchkiss Brain Institute, University of Calgary, Calgary, AB, Canada/Department of Surgery (Ophthalmology), University of Calgary, Calgary, AB, Canada.
¹⁰ Department Pediatrics, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada.
¹¹ Department of Ophthalmology and Vision Sciences, University of Toronto, Toronto, ON, Canada/Department of Ophthalmology and Vision Sciences, The Hospital for Sick Children, Toronto, ON, Canada.
¹² Division of Neurology, The Children's Hospital of Philadelphia, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
¹³ SickKids Research Institute, Neuroscience and Mental Health Program, The Hospital for Sick Children, Toronto, ON, Canada/Division of Neurology, The Hospital for Sick Children, Toronto, ON, Canada/Department of Pediatrics, University of Toronto, Toronto, ON, Canada.
¹⁴ Neuroscience and Mental Health Program, SickKids Research Institute, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada.

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 (n_eyes = 374) children with demyelinating diseases and 69 (n_eyes = 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.

Keywords: Multiple sclerosis; optical coherence tomography; pediatric; retinal nerve fiber layer thickness; supervised learning.

MeSH terms

Child
Humans
Machine Learning
Multiple Sclerosis* / diagnostic imaging
Retina / diagnostic imaging
Tomography, Optical Coherence*
Visual Pathways