Epileptic Seizure Detection by Quadratic Time-Frequency Distributions of Electroencephalogram signals

Abstract

Epilepsy is a chronic disorder caused by recurring seizures which can be detected by analyzing Electroencephalogram (EEG) signals. Several analysis methods have been deployed to recognize the non-stationary multicomponent content in relation with seizures within EEG signals. In this paper, we analyzed CHB-MIT Scalp EEG database signals by high-resolution quadratic time-frequency distributions, namely; the Spectrogram (SP), the Smoothed Pseudo Wigner-Ville Distribution (SPWVD), and the Choi-Williams Distribution (CWD) in a comparison viewpoint. We accomplished this study to evaluate the performance of each method towards detecting seizure within EEG signals. We used the time-frequency extended Renyi Entropy (RE) as a performance metric towards energy distribution complexity Within the time-frequency plane. We calculated this parameter over frequency bandwidths of Delta, Theta, Alpha, Beta, and Gamma brainwaves for the Quadratic Time-Frequency Distributions (QTFDs) of normal and abnormal EEG signals.