Analysis of EEG signals using empirical mode decomposition and support vector machine

Abstract

This paper presents a method for the detection of epileptic seizure from electroencephalogram (EEG) signals in the empirical mode decomposition (EMD) domain. Here we have used statistical moments like variance, skewness and kurtosis and non linear measures like sample entropy and approximate entropy on the intrinsic mode functions (IMFs), which are obtained by doing empirical mode decomposition on normal, interictal and ictal EEG signals. The intrinsic mode functions which are generated by the EMD method can be considered as a set of amplitude and frequency modulated signals. For classification the calculated features are feed to a support vector machine and the effectiveness of the proposed method is studied using a dataset which is available online. In the dataset we have considered three states namely normal, interictal and ictal. The proposed method gives an accuracy of 100% for the classification of normal and ictal as well as interictal and ictal EEG signals.