Continuous wavelet transform in the evaluation of stretch reflex responses from surface EMG

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Surface electromyography (sEMG) is crucial for analyzing muscle activity.
• Extracting dynamic reflex responses from sEMG, especially the tonic stretch reflex (SR), presents challenges due to low signal-to-noise ratios.
• Traditional methods like root mean square (RMS) have limitations in temporal resolution and accuracy.

Purpose of the Study:

• To introduce and validate the continuous wavelet transform (CWT) for analyzing muscle stretch reflexes from sEMG.
• To utilize a modulus-based method for estimating instantaneous amplitude-envelopes (sEMG intensity) from CWT ridges.
• To assess the method's efficacy in resolving both transient tendon reflexes and the tonic SR.

Main Methods:

• Eight neurologically healthy subjects underwent Achilles tendon taps and continuous ankle perturbation (0.1-8.0 Hz).
• sEMG data from soleus and tibialis anterior muscles were analyzed using both RMS and CWT-derived sEMG intensity.
• Reflex gain and coherence were calculated by cross-correlating sEMG estimates with the perturbation record.

Main Results:

• sEMG intensity correlated well with voluntary contraction torques.
• CWT successfully resolved M1, M2, and M3 reflex components with high accuracy and improved reconstruction compared to RMS.
• Wavelet ridge extraction yielded significantly higher coherence for the tonic SR than RMS, even in resting muscles.

Conclusions:

• Wavelet-ridge extraction is a valid and robust method for sEMG-based reflex evaluation.
• This approach is independent of absolute sEMG amplitude, facilitating inter-subject comparisons.
• It offers advantages for analyzing reflexes on an absolute frequency scale without normalization to maximal contractions.