Multiscale entropy analysis of different spontaneous motor unit discharge patterns

Area of Science:

• Biomedical Engineering
• Neuroscience
• Signal Processing

Background:

• Spontaneous electromyogram (EMG) signals exhibit complex patterns.
• Characterizing these patterns is crucial for understanding neuromuscular function.
• Existing methods may not fully capture the nonlinear dynamics of EMG signals.

Purpose of the Study:

• To apply multiscale entropy (MSE) analysis for characterizing diverse spontaneous EMG patterns.
• To compare the efficacy of standard MSE and intrinsic mode entropy (IMEn) analysis.
• To investigate the nonlinear dynamic properties of different spontaneous EMG signal types.

Main Methods:

• Application of standard Multiscale Entropy (MSE) analysis to spontaneous EMG.
• Utilized Intrinsic Mode Entropy (IMEn), based on multivariate empirical mode decomposition.
• Analyzed sporadic, tonic, and repetitive spontaneous motor unit discharges, and normal EMG baseline.

Main Results:

• Significant differences (p < 0.001) were found across multiple scales for both standard MSE and IMEn.
• Single-scale entropy analysis failed to reveal such distinctions.
• IMEn analysis demonstrated superior ability to differentiate EMG patterns using lower scale numbers compared to standard MSE.

Conclusions:

• MSE analysis provides a robust method for characterizing spontaneous EMG patterns.
• IMEn analysis offers enhanced sensitivity for discerning entropy differences in complex EMG signals.
• Findings enhance understanding of nonlinear dynamics in EMG, potentially relating to spinal motoneuron or motor unit health.