Evaluation of EEG signal during the A-phases of the cycling alternating pattern using principal component analysis

A-phases are transient events with a length between 2 and 60 s that can be observed in the EEG during NREM sleep. Recent studies have characterized the properties of the A-phases based on the spectral components of the EEG signal. These studies have given interesting insight about the Aphases and have helped to develop automatic algorithms for their detection. The characterization includes the evaluation of only the typical bands of the EEG signal, including Delta (0.03 Hz - 4 Hz), Theta (4 Hz - 8 Hz), Alpha (8 Hz - 12 Hz) and Beta (12 Hz - 30 Hz); however, to our knowledge, there are not any sleep EEG studies during the A-phases involving frequencies higher than the Beta band (e.g., the Gamma band (30 Hz - 100 Hz)). This paper focuses on the characterization of the spectral components of the A-phases on EEG activity by means of Principal Component Analysis, considering the full bandwidth of the EEG signal (up to 100 Hz). The results showed that frequency components above 50 Hz exist, especially during the A3-phase. Thus the results suggest that Gamma activity can be found during the NREM sleep process.