Statistical information approaches for the modelling of the epileptic brain

First, the theory of random process is linked with the statistical description of epileptic human brain process. A statistical information approach to the adaptive analysis of the electroencephalogram (EEG) is proposed. Then, the problem of time window recognition of the global stochastic model based upon Bayesian estimation and the use of global optimization for restricted experimental data are proposed. A robust algorithm for estimating unknown parameters of stochastic models is considered. The ability of nonlinear time-series analysis to extract features from brain EEG signal for detecting epileptic seizures is evaluated. (C) 2003 Elsevier Science B.V. All rights reserved.