Continuous Wavelet Transform study for normal and ictal eeg signals analysis

Abstract

In this article we propose the use of the continuous wavelet transform to analyze some specific signals acquired via electroencelalogram (EEG).  The EEG is a widely used test to analyze the brain electrical activity and, through it, it is possible to detect, in certain frequency bands, if this electrical activity is in accordance with the established norms. To achieve this goal, we will first explain a little about how the EEG works and then use the continuous wavelet transform, through a program written in python language, to analyze the behavior of EEG signals in the time and frequency domains. There are two specific types of signals studied: from healthy people and from people in epileptic seizures, known as ictal signals.  Signals from a consolidated database of the University of Bonn were used.  The main results were statistical data obtained from the signals and their respective scalograms, in which the difference between the maximum values, mean values, and other parameters of each signal and scalogram were compared.