Pervasive brain monitoring and data sharing based on multi-tier distributed computing and linked data technology.

Area of Science:

• Neuroscience
• Computer Science
• Wearable Technology

Background:

• Real-world Brain-Computer Interface (BCI) applications face challenges in wearable systems for reliable cognitive state prediction.
• Advances in sensors, wireless communication, and distributed computing offer solutions for wearable BCI development.

Purpose of the Study:

• To develop a pervasive, on-line EEG-BCI system integrating Fog and Cloud Computing, semantic search, and adaptive models.
• To demonstrate the feasibility of a multi-tier computing architecture for real-time EEG-BCI applications.

Main Methods:

• Utilized wireless dry-electrode EEG headsets and MEMS motion sensors for data acquisition.
• Implemented a multi-tier architecture: Android phones (UI), personal computers (Fog Servers), and NCHC clusters (Cloud Servers).
• Employed semantic Linked Data search and adaptive prediction/classification models.

Main Results:

• Successfully conducted synchronous multi-modal global data streaming.
• Demonstrated a multi-player on-line EEG-BCI game.
• Pilot systems are being adapted for real-life stress monitoring and Parkinson's disease patient monitoring.

Conclusions:

• The developed pervasive on-line EEG-BCI system, leveraging advanced computing and sensing, shows promise for real-world applications.
• Future work includes developing BCI ontology, semantic annotation, and progressive model refinement for enhanced functionality.