Automated Epileptic Seizure Detection in EEG Signals Using FastICA and Neural Network

Brain is one of the most vital organs of humans, controlling thecoordination of human muscles and nerves. The transient andunexpected electrical disturbances of the brain results in an acutedisease called Epileptic seizures. Epileptic seizures typically lead toan assortment of temporal changes in perception and behavior. Asignificant way for identifying and analyzing epileptic seizureactivity in humans is by using Electroencephalogram (EEG) signal. In a significant number of cases, detection of the epileptic EEGsignal is carried out manually by skilled professionals, who are smallin number. This necessitates automated epileptic seizure detectionusing EEG signals. To add with, a number of researchers havepresented automated computational methods for detecting epilepticseizures from EEG signals. In this article, we propose a novel andefficient approach for automatically detecting the presence ofepileptic seizures in EEG signals. First, the input EEG signals areanalyzed with the aid of Fast Independent Component Analysis (FastICA), a Statistical Signal Processing Technique, to obtain thecomponents related to the detection of epileptic seizures. TheBackPropagation Neural Network is trained with the obtainedcomponents for effective detection of epileptic seizures. Theexperimental results portray that the proposed approach efficientlydetects the presence of epileptic seizure in EEG signals and alsoshowed a reasonable accuracy in detection