Detection of Dyslexia Using EEG Signal Decompositional technique: Variational mode decomposition and derivatives

Abstract

Dyslexia is a common, specific learning disability that is influenced by neurobiological, genetic, and environmental factors, affecting individuals' ability to read and spell. Unlike its historical definition, DSM-5 classifies dyslexia as a type of neurodevelopmental disorder and emphasizes its hereditary and lifelong nature. The use of EEG signals instead of cognitive tests is explored for dyslexia assessment in this study. Processing EEG data can help identify changes in brain activity before and after treatment, providing valuable information about the effectiveness of interventions. This study compared three different Variational Mode Decomposition (VMD) algorithms to detect improvement in dyslexia using EEG data. The results showed that these algorithms could significantly contribute to determining improvement in dyslexia and that the VMD and Multivariate Variance Mode Decomposition (MVMD) algorithms worked more precisely and stable than the Successive Variance Mode Decomposition (SVMD) algorithm in diagnosing dyslexia and monitoring treatment. © 2024 Elsevier B.V., All rights reserved.