Classification of EEG Signal into Different Epileptic Stages based on Feature Extraction Srividhya G1,*, Farheen Syed Uzma2, Hemalatha R. J.1, Priya Sangeetha1, Rubi Jaya1 1Assistant Professor, Department of Biomedical Engineering, Vels Institute of Science Technology and Advanced Studies, Chennai, India 2Student, Department of Biomedical Engineering, Vels Institute of Science Technology and Advanced Studies, Chennai, India *Corresponding Author: Srividhya G, Assistant Professor, Department of Biomedical Engineering, Vels Institute of Science Technology and Advanced Studies, Chennai, India Email: srividhya.se@velsuniv.ac.in
Online published on 4 June, 2019. Abstract Epilepsy affects almost 50 million populations worldwide and it is one of the most common disorders of the neurological system. Epilepsy is characterised by chronic seizures with a broad range of seizure types based on the condition that differs from one person to another. There various diagnostic modes such as EEG, Magnetic Resonance Imaging (MRI) and f MRI. In this paper, we have made an approach to classify epileptic seizures of EEG signals with ease, instead of using complex and ancient methods of decision making and to describe and find out which wave has significant disturbance in the Epileptic condition. In the work carried out, firstly, the EEG signals were split into separate frequency band wave and time-frequency features were extracted. Then, the extracted features were fed to classifier system, the Linear SVM and ANN classifier and the performances between the classifier were compared in terms of Accuracy, Sensitivity and Specificity. Top Keywords Electroencephalogram (EEG), Epilepsy Linear SVM, ANN. Top |