ReSurfEMG: A Python Package for Comprehensive Analysis of Respiratory Surface EMG.

Area of Science:

• Biomedical Engineering
• Respiratory Physiology

Background:

• Mechanical ventilation in respiratory failure requires balancing muscle loading and gas exchange.
• Surface electromyography (sEMG) non-invasively monitors respiratory muscles but lacks standardized processing.
• Interplay between ventilator and respiratory muscles is crucial for effective ventilatory support.

Purpose of the Study:

• Introduce the open-source Python package, ReSurfEMG, for respiratory sEMG analysis.
• Address challenges in sEMG signal processing standardization and documentation.
• Evaluate the impact of filtering settings on sEMG features.

Main Methods:

• Developed ReSurfEMG, integrating denoising, feature extraction, and quality assessment.
• Compared over-filtering, under-filtering, and ReSurfEMG default settings.
• Assessed effects on waveform duration, time-to-peak, amplitude, ETP, pseudo-slope, pseudo-SNR, AUB, and bell-curve error.

Main Results:

• Under-filtering increased amplitude (+21%) and ETP (+10%).
• Over-filtering smoothed waveforms, reducing amplitude (-58%), ETP (-39%), and pseudo-slope (-49%).
• ReSurfEMG defaults yielded highest SNRs with comparable or lower AUB and bell-curve errors.

Conclusions:

• ReSurfEMG integrates advanced methods for dedicated respiratory sEMG analysis.
• Signal processing settings significantly impact sEMG features.
• ReSurfEMG promotes reproducible signal processing and standardization in respiratory sEMG analysis.