[Comparative study on evaluation algorithms for neck muscle fatigue based on surface electromyography signal]

Area of Science:

• Biomedical Engineering
• Ergonomics
• Physiology

Context:

• Neck muscle fatigue is a common issue during prolonged desk work.
• Quantitative evaluation of neck muscle fatigue is crucial for human factor analysis.
• Existing fatigue evaluation algorithms lack sufficient accuracy and reliability.

Purpose:

• To compare the effectiveness of various neck muscle fatigue evaluation algorithms.
• To identify a superior algorithm for quantitative assessment of cervical muscle fatigue.
• To evaluate algorithms based on anti-interference and fatigue distinguishing capabilities.

Summary:

• Surface electromyography signals from the sternocleidomastoid muscle were analyzed in 15 subjects using five algorithms: mean power frequency, spectral moments ratio, discrete wavelet transform, fuzzy approximation entropy, and complexity.
• The least squares method and Kolmogorov-Smirnov test were used to assess algorithm performance based on coefficient of determination (R²) and slope (k).
• Fuzzy approximation entropy demonstrated the highest R² for anti-interference and the largest value for distinguishing fatigue, outperforming other methods significantly (P < 0.05).

Impact:

• Fuzzy approximation entropy is identified as a more effective algorithm for evaluating cervical muscle fatigue.
• This finding can lead to improved human factor assessments and ergonomic interventions for desk workers.
• The study provides a validated quantitative method for neck muscle fatigue monitoring.