Does an EMD-based classification method using LS-SVM accurately classify focal and non-focal EEG signals?
An EMD-based feature extraction method combined with an LS-SVM classifier provides promising accuracy for classifying focal and non-focal EEG signals.
The electroencephalogram (EEG) signals are commonly used signals for detection of epileptic seizures. In this paper, we present a new method for classification of two classes of EEG signals namely focal and non-focal EEG signals. The proposed method uses the sample entropies and variances of the intrinsic mode functions (IMFs) obtained by empirical mode decomposition (EMD) of EEG signals. The average sample entropy (ASE) of IMFs and average variance of instantaneous frequencies (AVIF) of IMFs for separate EEG signals have been used as features for classification of focal and non-focal EEG signals. These two parameters have been used as an input feature set to the least square support vector machine (LS-SVM) classifier. The experimental results for various IMFs of focal and non-focal EEG signals have been included to show the effectiveness of the proposed method. The proposed method has provided promising classification accuracy for classification of focal and non-focal seizure EEG signals when radial basis function (RBF) has been employed as a kernel with LS-SVM classifier.
Sharma et al. (Thu,) studied this question.