Epileptic detection based on whale optimization enhanced support vector machine

• Published in: Journal of information & optimization sciences Vol. 40; no. 3; pp. 699 - 723
• Main Authors: Houssein, Essam H., Hamad, Asmaa, Hassanien, Aboul Ella, Fahmy, Aly A.
• Format: Journal Article
• Language: English
• Published: Taylor & Francis 03.04.2019
• Subjects: 68T and 68U; Discrete wavelet transform; Electroencephalogram (EEG); Epilepsy; Support vector machine; Whale optimization algorithm
• ISSN: 0252-2667; 2169-0103
• DOI: 10.1080/02522667.2018.1453671

A hybrid novel meta-heuristic optimization algorithm termed Whale Optimization Algorithm (WOA) and Support Vector Machines (SVMs) to obtain an Electroencephalogram (EEG) classification approach was proposed here in this paper for automatic seizure detection termed WOA-SVM. In the proposed approach, the Discrete Wavelet Transform (DWT) was applied to extract the main features and then decompose it into four level of decomposition tree. Furthermore, WOA was utilized to find the more significant feature subset of EEG from a larger feature pool as well as to enhance the parameters of SVM classifier. In order to detect epileptic, SVM with a Radial Basis kernel Function (RBF) was applied. Eventually, the proposed WOA-SVM approach is able to enhance the diagnosis of epilepsy as revealed from the statistical results with accuracy 100% for normal subject data versus epileptic data.