Sleep Dynamic Analysis Technology Based on Cross-Phase-Amplitude Transfer Entropy in Multiple Brain Regions

Area of Science:

• Neuroscience
• Computational Neuroscience
• Sleep Science

Background:

• Brain activity involves dynamic information flow across regions, varying significantly during sleep states.
• Phase-amplitude coupling (PAC) is established for assessing neural oscillation connectivity, but directional information flow remains underexplored.
• Understanding directional brain communication is crucial for characterizing complex sleep dynamics.

Purpose of the Study:

• To introduce a novel cross-phase-amplitude transfer entropy method for quantifying multi-regional sleep dynamics.
• To evaluate the effectiveness and accuracy of the proposed method using simulated and real sleep EEG data.
• To investigate directional information flow and connectivity patterns across different sleep stages.

Main Methods:

• Development of a cross-phase-amplitude transfer entropy algorithm to measure directional connectivity.
• Validation using simulations of multivariate nonlinear and nonstationary signals.
• Application to electroencephalogram (EEG) data from healthy adults during various sleep stages (Awake, N1, N2, N3, REM).

Main Results:

• Directional PAC flows from the occipital to the frontal lobe during Awake and N1 sleep stages.
• PAC directionality reverses to frontal-to-occipital in deeper sleep stages (N2, N3).
• The strength of delta-theta/alpha PAC increases with sleep depth, with REM sleep showing varied frequency-pair specific patterns.

Conclusions:

• The proposed cross-phase-amplitude transfer entropy method is a reliable tool for assessing brain function and connectivity during sleep.
• The study elucidates specific directional information flow patterns across different sleep stages.
• This framework offers potential for identifying and understanding multi-regional sleep dynamics and their clinical relevance.