Development of methodologies for the effective analysis of epileptic brain signals

Abstract

Epilepsy, one among a prevalent neurological condition affecting the central nervous system, creates mental or physical dilemmas in the affected patients. The traditional diagnosis is expensive and time-consuming. Thus, understanding the limitations that are significant of the existing methods especially computational complexities are addressed by this thesis and proposes novel and lightweight methods and features for rapid and accurate detection of epileptic seizures from the EEG signals which are all evaluated with multiple benchmark datasets, and the end results produced are promising in nature. The reduced computational complexity also enables the real time implementation of its automated detection using proposed methods and features in clinical systems and mobile EEG devices. newline