EEG-Based Functional Connectivity Representation using Phase Locking Value for Brain Network Based Applications.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Network neuroscience has advanced the study of brain networks using functional connectivity from multichannel EEG.
• Fully connected networks derived from EEG limit the characterization of complex brain networks.
• Significant Functional Networks (SFNs) offer a way to quantify and understand brain network dynamics.

Purpose of the Study:

• To present a framework for identifying SFNs corresponding to specific events from fully connected EEG networks.
• To enhance brain cognition analysis and brain network-based applications through event-related SFN identification.

Main Methods:

• Utilized phase-locking value (PLV) in EEG to determine event-specific PLV differences.
• Identified reactive frequency bands and most reactive pairs (MRPs) based on PLV differences.
• Constructed SFNs using the identified MRPs for specific events.

Main Results:

• The framework was applied to the DEAP dataset to identify SFNs associated with emotional states.
• Achieved comparable state-of-the-art accuracies in multi-class emotion classification using the identified SFNs.
• Demonstrated the method's efficacy as a general thresholding technique for event-related SFNs.

Conclusions:

• The proposed method effectively identifies event-related SFNs from EEG data.
• These SFNs are crucial for advancing brain network-based applications, particularly in emotion analysis.
• The framework provides a valuable tool for characterizing dynamic brain network activity.