Investigation of temporal variability of sleep EEG

In this study, a quantitative analysis method based on the local minima and maxima features is applied to the sleep EEG. The local minima (local maxima) is defined as a point whose amplitude is less (greater) than its neighbors. The average of local min-max intervals and the variance of local min-max intervals are used as temporal variability measures. From the computational results, it is observed that both temporal variability measures of the EEG tend to decrease with deeper sleep stages. This thus suggests that the EEG associated with a deeper sleep stage contains less amount of high frequency contents and has a wider range of frequency contents that the EEG associated with a lighter sleep stage. Furthermore, there are significant differences in both temporal variability measures of the EEG associated with different sleep stages.