|
Activity Number:
|
380
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303987 |
|
Title:
|
Multivariate Analysis of EEG Sleep Patterns of Neonates
|
|
Author(s):
|
Alexandra Piryatinska*+
|
|
Companies:
|
San Francisco State University
|
|
Address:
|
1600 Holloway Avenue, San Francisco, CA, 94002,
|
|
Keywords:
|
sleep stage separation ; time series ; nonstationary ; spectral analysis ; change point detection
|
|
Abstract:
|
We analyzed EEG-sleep patterns of the neonates. The level of brain dysmaturity is known to be directly related to the structure of neonatal sleep. In the past, the assessment of sleep EEG structure has often been done manually by an experienced clinician. We developed an algorithm for sleep stage separation using multichannel EEG signal. We separated different sleep stages corresponding to different stationary segments of the EEG signal based on statistical analysis of the spectral and nonlinear characteristics. The nonparametric change-point detection algorithm and cluster analysis were applied to these characteristics to obtain the sleep stage separation. We found the channels and characteristics which are most suitable to separate sleep stages.
|
