EEG signal classification using empirical mode decomposition and support vector machine

Abstract

In this paper, we present a new method based on empirical mode decomposition (EMD) for classification of seizure and seizure-free EEG signals. The EMD method decomposes the EEG signal into a set of narrow-band amplitude and frequency modulated (AM-FM) components known as intrinsic mode functions (IMFs). The method proposes the use of the area parameter and mean frequency estimation of IMFs in the classification of the seizure and seizure-free EEG signals. These parameters have been used as an input in least squares support vector machine (LS-SVM), which provides classification of seizure EEG signals from seizure-free EEG signals. The classification accuracy for classification of seizure and seizure-free EEG signals obtained by using proposed method is 98.33% for second IMF with radial basis function kernel of LS-SVM. © 2012 Springer India Pvt. Ltd.