High-density EEG coherence analysis using functional units applied to mental fatigue

Area of Science:

• Neuroscience
• Brain-Computer Interface
• Signal Processing

Background:

• Electroencephalography (EEG) coherence quantifies functional brain connectivity across frequencies.
• High-density EEG analysis presents challenges for traditional hypothesis-driven coherence methods due to numerous electrodes.
• Identifying relevant brain regions and coherences is crucial but difficult with conventional approaches.

Purpose of the Study:

• To apply a previously developed data-driven approach for analyzing high-density EEG coherence.
• To demonstrate the utility of this method in a case study of mental fatigue.
• To overcome limitations of conventional hypothesis-driven methods in EEG coherence analysis.

Main Methods:

• Utilized a data-driven approach to define regions of interest (ROIs) as functional units (FUs).
• FUs are spatially connected electrode sets with significantly coherent signals.
• Applied the method to analyze EEG coherence in a mental fatigue case study.

Main Results:

• The data-driven approach effectively identifies relevant coherences, overcoming limitations of hypothesis-driven methods.
• Visualization of group FU maps provides an economical summary of extensive EEG data.
• The method facilitates the selection of coherences for subsequent quantitative analysis.

Conclusions:

• The data-driven FU approach offers a powerful, efficient method for analyzing high-density EEG coherence.
• This technique complements traditional hypothesis-driven analyses by providing a data-driven basis for investigation.
• The approach is particularly beneficial for complex datasets and conditions like mental fatigue.