Development and Validation of a Low-Cost Open-Source Force Myography Band for Hand Gesture Recognition

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Human-Computer Interaction

Background:

• Wearable systems for hand gesture recognition are crucial for advanced control interfaces in prosthetics, exoskeletons, and virtual reality rehabilitation.
• Existing high-cost, proprietary systems limit accessibility in research and clinical settings.

Purpose of the Study:

• To develop and feasibility test a low-cost, open-source wearable forearm band for hand gesture recognition.
• To provide an accessible alternative to expensive, proprietary gesture recognition technologies.

Main Methods:

• Developed a wearable forearm band utilizing force myography (FMG) to detect radial muscle forces during hand gestures.
• Conducted a participant-based study (N=15) with able-bodied individuals performing 10 distinct hand gestures.
• Employed linear discriminant analysis for offline classification of FMG data.

Main Results:

• The FMG band achieved an average accuracy of 94.28% across all participants and gestures.
• Demonstrated comparable performance to existing, more expensive gesture recognition technologies.
• Feasibility of the low-cost, open-source system was confirmed.

Conclusions:

• The developed low-cost, open-source FMG band offers a viable and accurate solution for hand gesture recognition.
• This technology has the potential to enhance accessibility for research, clinical, and consumer applications.
• The system's performance supports its use in advanced control interfaces for various assistive and interactive technologies.