Nonlinear dynamic research on EEG signals in HAI experiment

摘要

According to the phase space reconstruction techniques from one-dimensional time series and the quantitative criteria of chaotic system, analyses and computations are conducted on Electroencephalogram (EEG) signals of piglets resulted from HAI (hypoxic–asphyxic injury) experiments. Based on the comparative studies of the phase graph, the power spectra, the correlation dimension, and the Lyapunov exponent of EEG signals of piglets, under normal and injury conditions, the following conclusions are shown: (1) the analyses of phase graph, power spectra, correlation dimension and Lyapunov exponent of EEG signals manifest the whole dynamic characteristics of cerebrum, and they may be used as new quantitative methods for early diagnosis of brain injuries. (2) Under normal physiological conditions, EEG signals are chaotic; while under pathologic conditions, EEG signals approach regularity.