Performance evaluation for epileptic electroencephalogram (EEG) detection by using Neyman-Pearson criteria and a support vector machine

► We extract EEG features by computing wavelet coefficients and approximate entropy. ► Neyman-Pearson criteria and an SVM are applied for feature-based detection. ► Simulation results show that these features represent the EEG signals well. ► An SVM applied on these features achieved higher detection accuracies than Neyman-Pearson criteria.