A Method for Epileptic Seizure Detection in EEG Signals Based on Tunable Q-Factor Wavelet Transform Method Using Grasshopper Optimization Algorithm With Support Vector Machine Classifier

 Epilepsy is a brain disorder disease that affects people’s quality of life. If it is detected at an early stage, seizures will not spread from the initial area. Electroencephalography (EEG) signals are used to diagnose epileptic seizures. However, this method cannot diagnose the state of epileptic seizure precisely. With the help of the Computer-Aided Diagnosis (CAD) system, neurologists can diagnose epileptic seizure stages correctly.
This study aims to present a novel method for epileptic seizures detection in EEG signals.

The Bonn dataset was used in this study with avaibale EEG signals divided into 5-second windows. Then, the Tunable Q-Factor Wavelet Transform (TQWT) was utilized to decompose the segmented EEG signals into various sub-bands. Several statistical and nonlinear features based on fractal dimension and entropy algorithms were extracted from the TQWT sub-bands. Then, the Autoencoder (AE) method with 7 layers was applied to reduce the number of features. Finally, the Support Vector Machine (SVM) and Grasshopper Optimization Algorithm with SVM classifier (GOA/SVM) were used for their classification compared to the K-Nearest Neighbors and Random Forest algorithms. The employment of AE for feature reduction and GOA/SVM for classification are the novelties of this study.

The proposed method demonstrated better performance compared to other methods used in different studies. The GOA/SVM classification method had a high accuracy rate of 99.42% and 99.23% for two-class and multi-class classification problems, respectively.

The combination of EEG feature classification methods increases the accuracy of the CAD system in diagnosing epileptic seizures. The method proposed in this study using different methods for extracting features and their classification has high accuracy for epileptic seizures detection.