Features extraction for classification of focal and non-focal EEG signals

Abstract

The electroencephalogram (EEG) is most often used signal to detect epileptic seizures. For a successful epilepsy surgery, it is very important to localize epileptogenic area. In this paper, a new method is proposed to classify focal and non-focal EEG signals. EEG signal is decomposed by empirical mode decomposition (EMD). The average Renyi entropy and the average negentropy of IMFs for EEG signals have been computed as features. The class discrimination ability of these features are quantified using Kruskal−Wallis statistical test. These features are set to input in neural network classifier for classification of focal and non-focal EEG signals. The experimental results are presented to show the effectiveness of the proposed method for classification of focal and non-focal EEG signals.