Surface electromyography for testing motor dysfunction in amyotrophic lateral sclerosis.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease affecting upper motor neurons (UMNs).
• Early and accurate diagnosis of UMN degeneration in ALS is crucial for patient management.
• Surface electromyography (sEMG) offers a non-invasive method to assess neuromuscular function.

Purpose of the Study:

• To evaluate the efficacy of dynamical features derived from sEMG signals for detecting UMN degeneration in ALS.
• To identify the most effective features and classification strategies for distinguishing ALS patients from healthy controls.

Main Methods:

• Acquired sEMG signals from upper limb muscles of 13 ALS patients and 20 controls.
• Extracted novel dynamical features characterizing temporal, frequency, and complexity aspects of muscle activity.
• Employed various classification algorithms, including decision trees, random forest, and Adaboost, to identify discriminating features.

Main Results:

• Significant differences in sEMG dynamical features were observed between ALS patients and controls.
• Classification accuracies reached up to 94%, particularly when analyzing forearm and hand recordings.
• Detrended fluctuation analysis and peak frequency emerged as robust discriminators, alongside specific machine learning classifiers.

Conclusions:

• Dynamical analysis of sEMG signals provides a viable method for identifying UMN changes in ALS.
• This approach demonstrates potential for non-invasive, accurate detection of ALS-related neuromuscular alterations.