Investigation of sleep stages identification with time-scale based parameters

In this study, the time-scale based analysis of EEG signals is shown for recognition of sleep stages. The EEG signals from healthy subjects are analyzed by Scalogram method in the time-scale domain. We observed that statistical parameters, the average gray level and measure of uniformity extracted from the energy distribution images, are found to be effective on the recognition of sleep stages.

[1]  Jose C. Principe,et al.  SAMICOS-A Sleep Analyzing Microcomputer System , 1986, IEEE Transactions on Biomedical Engineering.

[2]  M. Lehtokangas,et al.  Optimization of sigma amplitude threshold in sleep spindle detection , 2000, Journal of sleep research.

[3]  Jean Gotman,et al.  Computer-assisted sleep staging , 2001, IEEE Trans. Biomed. Eng..

[4]  Claudio A. Perez,et al.  Extracting Fuzzy Rules From Polysomnographic Recordings for Infant Sleep Classification , 2006, IEEE Transactions on Biomedical Engineering.

[5]  Saeid Sanei,et al.  EEG signal processing , 2000, Clinical Neurophysiology.

[6]  C. Robert Pinnegar,et al.  Time–Frequency Phase Analysis of Ictal EEG Recordings With the S-Transform , 2009, IEEE Transactions on Biomedical Engineering.