Assessing aberrant muscle activity patterns via the analysis of surface EMG data collected during a functional...

Area of Science:

• Biomedical Engineering
• Clinical Biomechanics
• Rehabilitation Science

Background:

• Surface electromyography (EMG) is crucial for assessing muscle activity in musculoskeletal disorders.
• Current visual inspection of EMG data is time-consuming and requires expert interpretation.
• Objective assessment of aberrant muscle activity patterns is needed for efficient diagnosis.

Purpose of the Study:

• To develop and validate algorithms for automatic evaluation of aberrant muscle activity patterns from surface EMG.
• To compare algorithmic estimates with expert-derived clinical scores.
• To assess the clinical utility of the automated technique in a real-world case study.

Main Methods:

• Developed a set of algorithms to automatically analyze EMG recordings from 62 subjects during functional evaluations.
• Generated clinical scores through expert visual inspection of EMG data on an ordinal scale.
• Utilized linear regression and Random Forest regression for algorithm development and comparison.
• Applied the algorithms to a case study of a patient with persistent back pain.

Main Results:

• EMG-based algorithms accurately estimated clinical scores, with a small root-mean-square error.
• Linear regression provided satisfactory results, with Random Forest used for comparison when needed.
• Regression coefficients confirmed a good fit between algorithmic estimates and expert scores.
• The algorithms successfully captured muscle activity patterns in the clinical case study.

Conclusions:

• The proposed EMG-based algorithms accurately estimate the severity of aberrant muscle activity.
• The technique offers a suitable method for deriving clinically relevant information from functional EMG data.
• Automated analysis of EMG data can enhance diagnostic efficiency and accuracy for musculoskeletal disorders.