Finger Joint Angle and Gesture Estimation Under Dynamic Hand Position Using a Soft Printed Electrode Array.

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Wearable Technology

Background:

• Surface electromyography (sEMG) shows promise for human-machine interfaces (HMI) but struggles with artifacts and variability.
• Existing sEMG gesture recognition often relies on static hand positions, limiting real-world application.
• Dynamic hand movements present significant challenges due to artifacts and individual differences.

Purpose of the Study:

• To develop an improved finger gesture recognition system using soft wearable sEMG sensors.
• To enhance finger joint angle prediction and gesture recognition across both static and dynamic hand positions.
• To address limitations of current sEMG-based systems, particularly in dynamic environments.

Main Methods:

• Integration of a soft wearable sEMG sensor with a Video-Vision-Transform model.
• Utilization of motion sensor-based training for enhanced data acquisition.
• Testing across diverse participants to assess variability and performance.

Main Results:

• Demonstrated excellent differentiation of finger angles and gestures despite inter-participant variability.
• Achieved high recognition accuracy: 0.85 for static and 0.87 for dynamic conditions in high-performing participants (N=16).
• Validated the system's effectiveness in both controlled and dynamic hand movement scenarios.

Conclusions:

• The developed system advances EMG-based gesture recognition for more robust real-world applications.
• This technology holds significant potential for neurorehabilitation and assistive devices.
• Contributes to next-generation wearable neurotechnology for individuals with neuromuscular disorders.