Linear and non-linear measures of the human neonatal EEG

Non-linear measures (Kolmogorov entropy, correlation dimension and a surrogate based “<f>z</f>” score) of the underlying structure of the human neonatal EEG were made and compared to linear measures (eigenvalues) and to EEGs obtained from older children (greater than 3 years of age). Strong age-dependent differences were noted for Kolmogorov entropy, the surrogate-based measure of non-linearity and eigenvalues. The eigenvalues decrease with age and the Kolmogorov entropy increases with age. The measure of global non-linearity decreased with age so that it was positive in only a small fraction of the greater than 3-year-old subjects. [Copyright &y& Elsevier]