Classification of EEG Signals using Nonlinear Features and Preprocessing Techniques

Saneesh Cleatus T,Dr Thungamani M

doi:10.35940/ijeat.e2789.0610521

Abstract

In this paper we study the effect of nonlinear preprocessing techniques in the classification of electroencephalogram (EEG) signals. These methods are used for classifying the EEG signals captured from epileptic seizure activity and brain tumor category. For the first category, preprocessing is carried out using elliptical filters, and statistical features such as Shannon entropy, mean, standard deviation, skewness and band power. K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) were used for the classification. For the brain tumor EEG signals, empirical mode decomposition is used as a pre-processing technique along with standard statistical features for the classification of normal and abnormal EEG signals. For epileptic signals we have achieved an average accuracy of 94% for a three-class classification and for brain tumor signals we have achieved a classification accuracy of 98% considering it as a two class problem.

Highlights

Electroencephalography (EEG) is a noninvasive technique used to record the electrical signal generated due to neural interactions in the brain
Objective of this work is to study the effect of different nonlinear techniques which is used in the preprocessing of EEG signals
The publicly available clinical data consisting of multichannel tumor EEG signals that are sampled at 250Hz using 16 bits per sample were taken for tumor detection

Summary

INTRODUCTION

Electroencephalography (EEG) is a noninvasive technique used to record the electrical signal generated due to neural interactions in the brain. The traditional filtering methods may not be enough to process and model as EEG signals are neither band-limited nor stationary [2],[3]. In the time and frequency domain approaches, traditional time domain methods like co-variances and correlations are generally be used. For the processing or pre-processing of EEG signals, standard linear transformation techniques like Singular Value Decomposition (SVD) and its related variations Principal Component Analysis (PCA), and Independent Component Analysis (ICA) are widely used [4]. Statistical models are widely used if a signal is said to nonlinear in nature as well as not stationary. For modelling non-stationary time series, ARMA is generalized to ARIMA (Auto Regressive Integrated Moving Average) models. ARIMA processes reduce to ARMA when the data is fitted to the difference data [5]

METHODOLOGY

Analysis of epileptic EEG signals

Analysis of EEG signals of brain tumor data

Results of proposed method for tumor affected brain EEG signals

Findings

CONCLUSION