Static hand posture classification based on the biceps brachii muscle synergy features

Area of Science:

• Biomedical Engineering
• Neuroprosthetics
• Human Motor Control

Background:

• Modern upper limb myoelectric prostheses require numerous control signals for complex movements.
• The biceps brachii muscle possesses multiple individually innervated compartments, suggesting potential for diverse signal generation.
• Understanding muscle synergies is crucial for decoding motor intent from electromyographic (EMG) signals.

Purpose of the Study:

• To investigate the feasibility of extracting muscular synergies solely from the biceps brachii muscle.
• To determine if these extracted synergies can be used to generate control signals for myoelectric prostheses.
• To assess the accuracy of classifying static hand postures based on biceps brachii EMG signals.

Main Methods:

• Applied non-negative matrix factorization (NMF) to analyze surface EMG signals.
• Collected EMG data from 8 electrodes placed on the biceps brachii of 10 healthy subjects.
• Recorded signals during static hand posture tasks in seated or standing positions.

Main Results:

• Muscular synergies were successfully extracted from the electrical activity of the biceps brachii.
• A learning process combined with a classifier enabled accurate identification of distinct hand postures.
• The system achieved a mean accuracy of over 80% in correctly detecting hand postures across subjects.

Conclusions:

• The biceps brachii muscle contains identifiable muscular synergies that can be exploited for prosthetic control.
• This approach offers a potential method for increasing the number of control signals available for upper limb prostheses.
• Further research could integrate this single-muscle synergy extraction into advanced myoelectric control systems.