Associating Dynamical Descriptors to EEG from Healthy and Epileptic Subjects

A.C. Ramos and H.G. González-Hernández (Mexico)

Keywords

Time Series Analysis, Attractor Reconstruction, Correla tion Dimension, Lyapunov Exponent, EEG, Epilepsy.

Abstract

In this work, there were analyzed and compared some dynamical properties of electroencephalogram (EEG) time series measured from epileptic subjects exhibiting generalized tonic-clonic seizures and epileptic background activity, and healthy subjects relaxed with open and closed eyes. First, the delayed coordinates technique was used to reconstruct the state space dynamics of each EEG segment. Then the correlation dimension and the largest Lyapunov exponent of the reconstructed attractors were computed. There were found significant differences between the dy namical descriptors of the analyzed EEG classes that make us believe that the present theory could be applied on a first stage of the diagnosis of epilepsy.

Important Links:



Go Back