Measuring phase-amplitude coupling using dispersion fuzzy mutual information

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Mild Cognitive Impairment (MCI) is an early stage of Alzheimer's disease (AD).
• Electroencephalography (EEG) signals and Phase-Amplitude Coupling (PAC) are promising for MCI detection.
• Existing PAC estimators have limitations in clinical application.

Purpose of the Study:

• To propose and evaluate the Dispersion Fuzzy Mutual Information (DFMI) method as a novel PAC estimator.
• To assess DFMI's applicability and performance compared to conventional PAC estimators.

Main Methods:

• The DFMI method integrates dispersion entropy and mutual information for time series analysis.
• It upgrades fuzzy entropy to a dual-channel algorithm, addressing pattern quantification challenges.
• DFMI performance was evaluated via simulations, comparing sensitivity, data length dependency, noise, and artifact resistance.

Main Results:

• DFMI effectively estimates PAC strength with reduced data length dependency and stable results.
• The method demonstrated strong resistance to pseudo-trace signals and artifacts.
• Analysis of MCI-EEG data revealed enhanced theta-gamma coupling and a shift in alpha-gamma coupling in MCI patients.

Conclusions:

• DFMI is a viable PAC estimator for neural signals.
• PAC alterations in MCI brains may present as coupling band attenuation.
• DFMI shows potential for improved early diagnosis of MCI using EEG.