Abstract

Neurological disorders pose a significant global health challenge, demanding efficient and accurate diagnostic tools. We Proposed a predictive framework for detecting neurological disorders based on brain connectivity indices derived from Electroencephalogram (EEG) signals. Leveraging brain connectivity indices as features, we develop a predictive model to classify common neurological disorders including Alcohol usage, Anxiety, Depression, and Schizophrenia. The framework integrates pre-processing steps to enhance the quality of EEG data and derive connectivity indices using Pearson Correlation Coefficient (PCC), Phase Locking Value (PLV) and Phase Lag Index (PLI) for identifying Brain Connectivity Patterns. We explore the effectiveness of Convolutional Neural Networks (CNNs) and their architectures in analysing EEG-derived features for disorder classification. Furthermore, we introduce a fusion model that combines the strengths of CNNs with additional methodologies to improve classification accuracy. Our fusion model enhances the robustness and performance of the predictive framework, offering promising results for automatic diagnosis systems aimed at assisting clinicians in early detection and intervention of neurological disorders.