Time domain analysis of EEG signals for detection of epileptic seizure

Abstract

The human brain produces electrical signals which prove vital in understanding the degree of abnormality that may, in many cases, result in a person behaving unusually. The information contained in these signals is recorded via an EEG machine, which is able to extract even the most subtle details from the electrical waves that the brain signals generate. This paper presents an autonomous system, capable of detecting the occurrence of a epileptic seizure, without the help of an expert. The proposed system consists of four steps i.e. preprocessing, where the data is organized in an orderly manner and noise is removed, followed by different time — domain features from the EEG recordings of a wide — range of epileptic patients, suffering with short, but random intervals of variable seizure intensity episodes. The system then performs feature selection to extract best set of features which are finally used for classification of EEG signals as normal or abnormal. The proposed system is tested on a publicly available dataset and has achieved an average accuracy of 82.5%.

Keywords Electroencephalography, Feature extraction, Databases, Scalp, Entropy