Development of an Algorithm for Extracting Limb Movement Candidates Using Surface Electromyography.

Area of Science:

• Sleep Medicine
• Biomedical Engineering
• Signal Processing

Background:

• Periodic limb movement disorder (PLMD) necessitates multi-night monitoring for accurate diagnosis and medication adjustment.
• Current polysomnography (PSG) is resource-intensive, making home monitoring essential for PLMD assessment.
• Surface electromyography (sEMG) is crucial for detecting limb movements (LMs) but requires robust algorithms for daily life settings.

Purpose of the Study:

• To develop and evaluate a novel four-step model for detecting periodic limb movements (PLMs) from sEMG data acquired in non-clinical environments.
• To create a preprocessing method and an amplitude-independent algorithm for extracting limb movement candidates.
• To establish a foundation for reliable PLMs home monitoring systems.

Main Methods:

• A four-step detection model was proposed, including preprocessing and an amplitude-independent LM candidate extraction algorithm.
• The algorithm was evaluated using sEMG data from 20 individuals diagnosed with PLMD, measured via PSG.
• Performance metrics focused on sensitivity and precision for identifying potential PLMs.

Main Results:

• The proposed algorithm achieved 96.7% sensitivity and 39.2% precision in extracting LM candidates from PSG-measured sEMG.
• This performance meets the physician-set sensitivity goal of 85% for the initial detection steps.
• The results indicate the algorithm's suitability for the initial stages of a PLMs home monitoring system.

Conclusions:

• The developed four-step model provides a viable approach for detecting PLMs from sEMG in daily life settings.
• This technology supports the development of practical PLMs home monitoring, overcoming PSG limitations.
• The study contributes to improved management of periodic limb movement disorder through accessible monitoring solutions.