Ali Mehravar - Department of Medical Bioengineering, Faculty of Advanced Medical Sciences, Tabriz University of MedicalSciences, Tabriz, Iran .Student Research Committee, Tabriz University of Medical Sciences, Tabriz, Iran
Samaneh Hoseini - Neurosciences Research Center, Tabriz University of Medical Sciences, Tabriz, Iran.
Yashar Amizadeh - Ophthalmic Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Seyed Hossein Rasta - Department of Medical Physics, Tabriz University of Medical Sciences, Tabriz, Iran. School of Medical Sciences, University of Aberdeen, Aberdeen, UK.

Introduction: Multiple Sclerosis (MS) is a prevalent autoimmune and inflammatorydisorder that leaves demyelination and neurodegenerative changes in Central NervousSystem (CNS). The retina is among the body organs that are affected by MS, OpticalCoherence Tomography (OCT) images can play a key role in the preliminary stages.Artificial intelligence-based methods are commonly applied in image classification andhave shown promising and applicable results in MS diagnosis.Method: In total, more than ۱۹۷ MS patients and more than ۲۸۳ healthy patients wereincluded in this study, and Spectralis OCT images were taken, and artificial intelligencewas trained with several thousand images using data augmentation. Finally, theautomatic diagnosis algorithm of MS disease was implemented, and then the diagramof network loss processes was drawn and the sensitivity, specificity and accuracy of thealgorithm were evaluated.Result: The disease was successfully diagnosed by OCT images with an accuracy of۹۳.۰.Conclusion: The proposed method showed improvements in early-stage MS diagnosis and with the potentiality to be used in either the diagnosis or prediction of the progression of other diseases that affect the CNS (e.g. Alzheimer's disease, bipolar disorder, etc.).

convolutional neural network, multiple sclerosis, optical coherence tomography, retinal layer thickness