Classification of seizure and seizure-free EEG signals based on empirical wavelet transform and phase space reconstruction

TL;DR

This paper presents a novel method combining empirical wavelet transform and phase space reconstruction to classify seizure and seizure-free EEG signals with high accuracy, outperforming previous techniques.

Contribution

The study introduces a new approach using EWT and phase space features for EEG classification, achieving 98% accuracy in seizure detection.

Findings

Achieved 98% classification accuracy

Used phase space features from EEG rhythms

Outperformed previous methods

Abstract

Epilepsy is a brain disorder due to abnormalactivity of neurons and recording of seizures is of primary interest in the evaluation of epileptic patients. A seizureis the phenomenon of rhythmicity discharge from either a local area or the whole brain and the individual behavior usually lasts from seconds to minutes.In this work, empirical wavelet transform(EWT) is applied to decompose signals into Electroencephalography (EEG) rhythms. EEG signals are separated to delta, theta, alpha, beta and gamma rhythms using EWT.The proposed method has been evaluated by benchmark dataset which is freely downloadable from Bonn University website. 95% confident ellipse area is computed from 2D projection of reconstructed phase space (RPS)of rhythms as features and fed to K-nearest neighbor classifier for detection of seizure (S) and seizure free (SF) EEG signals. Our proposed method archived 98% accuracy in classification of S and SF EEG signals with a tenfold cross-validation strategy that is higher than previous techniques.