Automatic epileptic seizures detection and eeg signals classification based on multi-domain feature extraction and multiscale entropy analysis

Abstract

Epilepsy is a frequently observed neurological abnormality. In the manual method, a physician monitors the recording of Electroencephalogram (EEG) of a patient to detect epileptic seizures. But this method is time-consuming and fallible. This chapter presents an automatic epileptic seizures detection and EEG signals classification method based on multi-domain feature extraction and multiscale entropy analysis. In this method, EEG data is collected and preprocessed for artifacts removal from the original data. Then discrete Fourier transform (DFT) and discrete wavelet transform (DWT) are applied for extracting features. Multiscale entropy (MSE) with sample entropy is also applied to extract nonlinear features. Finally, supervised learning classifiers such as support vector machine (SVM), k-nearest-neighbors (k-NN) and artificial neural network (ANN) are used for epileptic seizures detection, three-class and five-class classification of the EEG signals. The dataset has been collected from the University of Bonn. This method attained an accuracy of 99.54%, sensitivity of 98.56% and specificity of 99.76% for epileptic seizures detection. For three-class and five-class classification, accuracy was 98.22% and 87.00%, respectively.