Bessel k-form parameters in the dual tree complex wavelet transform domain for the detection of epilepsy and seizure

Abstract

In this paper, a statistical analysis of EEG signals is carried out in the dual tree complex wavelet transform (DT-CWT) domain. It is shown that Bessel k-form(BKF) pdf can suitably model the DT-CWT sub-bands and the BKF parameters in various DT-CWT sub-bands can discriminate various types of EEG data effectively. Next these parameters are utilized by the SVM-based classifiers to classify the EEG data. The classification performance is studied for three clinically relevant cases including healthy vs seizure, non-seizure vs seizure and inter-ictal vs ictal recordings. The proposed method provides 100% accuracy with 100% sensitivity and 100% specificity in all the cases. In addition, in comparison to several state-of-the-art algorithms, the proposed method has also been shown to be computationally fast.