Tongue-rudder: a glossokinetic-potential-based tongue-machine interface

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Glossokinetic potentials (GKPs) are electric signals generated by tongue movements.
• Existing brain-computer interfaces often face challenges with invasiveness and fine control.

Purpose of the Study:

• To develop a non-invasive tongue-machine interface for detecting tongue position and direction.
• To apply this interface for controlling external devices, specifically a wheelchair.

Main Methods:

• Utilizing a specific electrode configuration to capture discriminative GKPs.
• Developing a linear model to decode tongue direction from GKPs, addressing baseline drift.
• Integrating an electromyography-based system for teeth clenching detection as a stop command.

Main Results:

• Demonstrated that GKPs vary predictably with tongue direction.
• Successfully developed a linear model for robust tongue direction estimation.
• Validated the tongue-machine interface, termed 'tongue-rudder', for both offline cursor control and online wheelchair navigation.

Conclusions:

• The developed tongue-machine interface offers non-invasiveness, fine controllability, and compatibility with other EEG-based systems.
• GKPs provide a promising avenue for intuitive and effective human-machine interaction.
• The tongue-rudder system represents a significant advancement in assistive technology for wheelchair users.