Discrimination of Focal and Non-Focal Epileptic Eeg Signals Using Different Types of Classifiers

Abstract Epilepsy is a neurological disorder characterized by recurrent seizures and has high incidence rate. The aim of this research to classify EEG signals as either focal non-focal in order identify the epileptogenic area brain, which can be surgically treated manage epilepsy. In paper, was proposed classification method based on higher spectra (HOS) parameters four different classifiers: linear discriminant analysis (LDA), quadratic (QDA), k-Nearest neighbors (KNN), Mahalanobis distance (MD). evaluated using public dataset that consists recordings from epileptic patients. classifiers performances were it shown KNN classifier achieves maximum rate 99.55%, sensitivity 100%, specificity 99.09%. data performed with values 0.96 for F1-score, 0.91 both Kappa Matthews Coefficient. results demonstrate efficiency type signals.