Classification of Seizures in EEG Using Wavelet-Chaos Methodology and Genetic Algorithm

Abstract

Detection of seizures in EEG can be challenging because of myogenic artifacts and might be time-consuming when reviewing long term EEG recordings. In this study, we propose a method based on wavelet-chaos methodology and genetic algorithm for automatic seizure detection in EEG. The data used in this research are obtained from both healthy and epileptic subjects and are available online from the University of Bonn, Germany. The wavelet packet transform was used to decompose EEG into five EEG subbands: delta, theta, alpha, beta, and gamma. Non-linear parameters, including the time lag, the embedding dimension, the correlation dimension, and the largest Lyapunov exponent, were extracted from each of the frequency band and the original EEG signals. The nonlinear parameters were employed as the features to train the support vector machine (SVM) classifier. Finally, a genetic algorithm (GA) was used for selecting effective feature subset for the SVM classifier. Three groups of EEG recordings, including the EEG from healthy subjects, and the interictal and the ictal EEGs from epileptic subjects, were recruited for the study. When EEG recordings were classified into seizure free (A+B) and seizure (C) groups by SVM, the accuracies were 90% and 81.2%, respectively. The accuracies increased to 93.3% and 91.8%, respectively, by using SVM combined with GA. When EEG recordings were classified into three groups (A, B, and C) by SVM, the accuracies were 84%, 76.4% and 78.2%, respectively. And, the accuracies increased to 91.4%, 85% and 86.6%, respectively, by using SVM combined with GA. The results demonstrate that the SVM classifier with nonlinear features is effective in seizure classification in EEG. The performance was improved when the GA-based optimal feature subset selection method was employed.