Could the beta rebound in the EEG be suitable to realize a "brain switch"?

Area of Science:

• Neuroscience
• Brain-Computer Interfaces
• Signal Processing

Background:

• Foot motor imagery elicits event-related desynchronization (ERD) and a subsequent beta event-related synchronization (ERS), also known as beta rebound.
• The beta rebound occurs after the motor imagery task is completed.

Purpose of the Study:

• To investigate the suitability of the post-imagery beta rebound as a feature for a simple brain switch.
• To determine if the beta rebound can reliably detect a specific brain state for brain-computer interface (BCI) applications.

Main Methods:

• Electroencephalography (EEG) data, specifically Laplacian derivations at the vertex (Cz), were recorded from 5 healthy subjects during cue-based foot motor imagery.
• Event-related desynchronization (ERD) and beta rebound (ERS) were analyzed between 6 and 40Hz.
• Support vector machines (SVMs) were employed for classification of EEG patterns.

Main Results:

• The electroencephalography (EEG) signal showed a true positive rate (TPR) of 59.2% for detecting the beta rebound, significantly higher than the ERD (28.4%).
• In single runs with 30 cues, the TPR for beta rebound detection increased to 78.6%.
• The false positive rate for beta rebound detection was consistently maintained below 10%.

Conclusions:

• The beta rebound at Cz during foot motor imagery is a stable and reproducible EEG phenomenon.
• The findings indicate that the beta rebound is a suitable feature for developing a brain switch using a single EEG channel.