Optimizing Onset and Offset Detection in Surface Diaphragm Electromyographic Signals: A Signal Quality-Driven...

Area of Science:

• Respiratory Physiology
• Biomedical Engineering
• Pulmonology

Background:

• Chronic Obstructive Pulmonary Disease (COPD) management often involves nocturnal non-invasive mechanical ventilation (NIV).
• Accurate monitoring of diaphragm electromyographic (EMG) activity is crucial for assessing respiratory effort and ventilation effectiveness in COPD patients.
• Current methods for analyzing diaphragm EMG onset and offset can be time-consuming and cumbersome.

Purpose of the Study:

• To develop and validate an improved automatic algorithm for detecting the onset and offset of diaphragm EMG activity.
• To enhance the accuracy and efficiency of respiratory pattern analysis in COPD patients undergoing nocturnal NIV.
• To provide a more reliable tool for monitoring diaphragm function during NIV.

Main Methods:

• A signal quality-driven approach was implemented to refine an automatic algorithm for diaphragm EMG onset and offset detection.
• The algorithm was tested on diaphragm EMG data from patients with COPD receiving nocturnal NIV.
• Performance was evaluated by comparing automatic detection with manual scoring, focusing on accuracy improvements.

Main Results:

• The improved automatic detection algorithm significantly enhanced the accuracy of diaphragm EMG onset and offset detection.
• Average onset detection error was reduced from 194 ms to 29 ms.
• Average offset detection error was reduced from 285 ms to 119 ms.

Conclusions:

• The developed signal quality-driven approach effectively improves the accuracy of automatic diaphragm EMG onset and offset detection.
• This enhanced algorithm offers a valuable tool for more precise monitoring of respiratory mechanics in COPD patients on nocturnal NIV.
• The findings suggest potential for improved patient management and therapeutic adjustments based on more reliable EMG data.