Epileptic Seizure Detection Using Empirical Mode Decomposition

Abstract

In this paper, we attempt to analyze the performance of the Empirical Mode Decomposition (EMD) for discriminating epileptic seizure data from the normal data. The Empirical Mode Decomposition (EMD) is a general signal processing method for analyzing nonlinear and nonstationary time series. The main idea of EMD is to decompose a time series into a finite and often small number of intrinsic mode functions (IMFs). EMD is an adaptive decomposition since the extracted information is obtained directly from the original signal. By utilizing this method to obtain the features of normal and epileptic seizure signals, we compare them with traditional features such as wavelet coefficients through two classifiers. Our results confirmed that our proposed features could potentially be used to distinguish normal from seizure data with success rate up to 95.42%.