[Analyzing sleep EEG using correlation dimension and approximate entropy].

Abstract

Correct sleep scoring is the base of sleep studying; nonlinear features of EEG can represent different sleep stages. In this paper, correlation dimension (D2) and approximate entropy (ApEn) of sleep EEG have been calculated. The statistical results reveal that: D2 does not come to be saturated when the embedding dimension increases, but the relative value of D2 can effectively distinguish different sleep stages. ApEn has the advantage of calculating simply, steady result and representing preferably different sleep stages. ApEn and the relative value of D2 reveal, from different point of view, the same rule about EEG (brain) complexity changing, that is, both complexity and its fluctuation are maximal in the subject's awake hour, are decreasing with the deepening of sleep, but the complexity in REM is about the level between S1 and S2.