Classification Of Seizure And Seizure-Free Eeg Signals Using Hjorth Parameters

In this work, we used flexible analytic wavelet transform (FAWT) for the decomposition of electroencephalogram (EEG) for the the analysis of epileptic seizure in EEG signals with Hjorth parameters as features for these signals. For the classification of EEG signals, the chosen classifiers are twin support vector machines, least squares twin support vector machines and robust energy-based twin support vector machines for seizure and seizure-free signals. We apply 10-fold cross-validation to ensure the reliability of the results and to avoid over-fitting of the model. The maximum accuracy achieved in this work is 98.33%. Our proposed approach is found to be comparable with other baseline approaches present in the literature.