Sensor Level Functional Connectivity Topography Comparison Between Different References Based EEG and MEG.

Area of Science:

• Neuroscience
• Brain Network Analysis
• Biomedical Engineering

Background:

• Sensor-level functional connectivity topography (sFCT) is crucial for understanding brain networks.
• Both electroencephalography (EEG) and magnetoencephalography (MEG) can be used to construct sFCT.
• The choice of EEG reference can influence sFCT results.

Purpose of the Study:

• To compare sFCT within the EEG modality using different reference schemes.
• To compare EEG-based sFCT with MEG-based sFCT.
• To evaluate the impact of EEG reference techniques on inter-modality sFCT similarity.

Main Methods:

• Simulations were used to assess EEG references (REST, AR, LM, LR) and their effect on EEG-sFCT.
• Inter-modality comparisons between EEG-sFCT and MEG-sFCT were performed using Relative Error (RE), Overlap Rate (OR), and Hamming Distance (HD).
• Simultaneously recorded MEG and EEG data from face-recognition experiments were analyzed.

Main Results:

• The Reference Electrode Standardization Technique (REST) minimized reference effects on EEG-sFCT, yielding results closer to ground truth.
• REST-based EEG-sFCT demonstrated the highest similarity with MEG-sFCT across metrics (RE, OR, HD), particularly when data were recorded simultaneously.
• Analysis of real data confirmed REST-based sFCT closely matched MEG-sFCT.

Conclusions:

• The choice of EEG reference significantly impacts sFCT.
• REST is recommended for EEG-based sFCT to enhance comparability with MEG-sFCT.
• This study provides valuable insights into MEG-EEG sFCT similarity and reference scheme selection.