An isometric muscle force estimation framework based on a high-density surface EMG array and an NMF algorithm.

Area of Science:

• Biomedical Engineering
• Neuroscience
• Kinesiology

Background:

• Accurate muscle force estimation is crucial for understanding muscle function and developing advanced prosthetic devices.
• Traditional methods often struggle with the complexity and heterogeneity of muscle activation patterns.

Purpose of the Study:

• To develop a novel framework for precise muscle force estimation.
• To identify and extract relevant muscle activation signals from specific areas for improved prediction accuracy.

Main Methods:

• Utilized high-density (HD) surface electromyographic (sEMG) grids to record biceps brachii activity during isometric contractions.
• Employed nonnegative matrix factorization (NMF) to decompose sEMG signals and identify major muscle activation patterns.
• Selected channels with high weighting factors corresponding to the dominant activation pattern for force estimation using polynomial fitting.

Main Results:

• The proposed method significantly enhanced the accuracy of muscle force estimation.
• A reduction in the number of required electrodes was achieved compared to conventional approaches using the entire grid.
• The identified major activation pattern effectively represented the muscle's response during force generation.

Conclusions:

• The novel framework offers an effective strategy for precise muscle force estimation by focusing on optimal electrode placement.
• This approach has potential applications in muscle heterogeneity analysis, myoelectric prostheses control, and exoskeleton device operation.
• The findings contribute to advancing non-invasive methods for assessing neuromuscular function.