Real-time modeling and 3D visualization of source dynamics and connectivity using wearable EEG

Area of Science:

• Neuroscience
• Biomedical Engineering
• Data Science

Background:

• Real-time analysis of electroencephalography (EEG) data is crucial for advancing brain-computer interfaces and neurological studies.
• Existing toolboxes often require complex integration for comprehensive EEG data processing pipelines.

Purpose of the Study:

• To develop and validate an integrated, open-source pipeline for real-time EEG data extraction, preprocessing, artifact rejection, source reconstruction, and analysis.
• To demonstrate the pipeline's efficacy using simulated data and novel high-density dry EEG recordings.

Main Methods:

• Utilized the open-source SIFT and BCILAB toolboxes for a comprehensive EEG analysis workflow.
• Implemented real-time data extraction, preprocessing, artifact rejection, source reconstruction, and multivariate dynamical system analysis (including spectral Granger causality).
• Applied 3D visualization and classification techniques to processed data.

Main Results:

• Successfully demonstrated real-time data processing capabilities for EEG signals.
• Validated the pipeline's performance on both simulated datasets and data from a new 64-channel wearable dry EEG system.
• Achieved effective artifact rejection, source reconstruction, and multivariate analysis.

Conclusions:

• The developed pipeline offers a robust and efficient solution for real-time EEG analysis.
• The integration within SIFT and BCILAB facilitates broader accessibility and application in neuroscience research.
• The system shows promise for applications requiring immediate EEG data interpretation, such as advanced BCI systems.