Multifractal Analysis and Wavelet Decomposition in EEG Signal Classification

Abstract

Multifractal measures have shown significant differences among normal, interictal, and ictal EEGs. The proposed scheme is demonstrated with high accuracy through suitable graphical methods and statistical tools called the one-way Analysis of Variance (ANOVA) test with a Box Plot. It is shown that the designed methods perform significantly in the detection of epileptic seizures in EEG signals. The EEG data are further tested for linearity by using the Normal Probability Plot and have proved that Epileptic EEG has significant nonlinearity whereas Healthy EEG is distributed normally and similar to the Gaussian Linear Process.