Effect of surface electromyography normalisation methods over gait muscle synergies

Area of Science:

• Biomedical Engineering
• Neuroscience
• Human Movement Science

Background:

• Muscle synergies are fundamental to understanding motor control.
• Electromyography (EMG) data processing requires appropriate normalization for accurate analysis.
• Variability in normalization techniques can impact muscle synergy extraction.

Purpose of the Study:

• To investigate the impact of six different normalization methods on muscle synergy extraction from EMG data.
• To assess how normalization affects synergy complexity, structure, and activation coefficients during walking in children.
• To identify the most reliable normalization methods for pediatric motor control research.

Main Methods:

• EMG data were collected from 18 typically developing children (aged 8-15) during walking trials.
• Six normalization methods were applied: Raw, Within-Trial Maximum (WTM), Inter-Trial Maximum (ITM), Task-Specific Maximum (TSM), Magnitude Percentile (MP), and Unit Variance.
• Non-negative matrix factorization (NMF) was used to extract muscle synergies.

Main Results:

• Normalization method significantly influenced the number, structure, and temporal characteristics of extracted muscle synergies.
• TSM and ITM methods resulted in more consistent muscle synergies compared to other methods.
• MP and WTM methods demonstrated greater variability in synergy extraction.

Conclusions:

• The choice of normalization method critically affects muscle synergy analysis in pediatric populations.
• TSM and ITM are recommended for more robust and reliable muscle synergy extraction during walking.
• Standardizing normalization procedures is essential for advancing research in motor control and developing unified processing workflows.