Time-variant Parametric Estimation of Transient Quadratic Phase Couplings during Electroencephalographic Burst Activity

摘要: Objectives: Electroencephalographic burst activity characteristic of burst-suppression pattern (BSP) in sedated patients and burst-interburst (BIP) the quiet sleep healthy neonates have similar linear non-linear signal properties. Strong interrelations between a slow frequency component rhythmic, spindle-like activities with higher frequencies been identified previous studies. Time-varying characteristics BSP BIP prevent definite patternrelated analysis. A continuous estimation bispectrum is essential to analyze these patterns. Parametric bispectral approaches provide this opportunity. Methods: The adaptation an AR model leads parametric by using transfer function estimated filter. Time-variant require parameters which consider order moments preserve phase information. Accordingly, time-variant was introduced. Data driven simulations were performed optimal parameters. (12 patients) (6 neonates) analyzed novel approach. Results: Significant differences time course during burst-like before onset could be shown. rhythmic quadratic coupling (period 10 sec) all neonates. Conclusion: Quadratic couplings increases depending on depth sedation. visually detected only temporarily observable EEG correlate hidden neural process. offer possibility better characterization underlying processes leading improved diagnostic tools used clinical routine.