Ambulatory respiratory rate detection using ECG and a triaxial accelerometer.

Area of Science:

• Biomedical Engineering
• Physiological Monitoring
• Wearable Technology

Background:

• Continuous respiratory rate monitoring is crucial for detecting respiratory diseases and enabling remote patient care.
• Existing unobtrusive monitoring techniques often lack sufficient accuracy.
• Developing accurate and low-complexity algorithms for wearable sensors is a key challenge.

Purpose of the Study:

• To develop and evaluate a computationally efficient algorithm for estimating respiratory rate using an unobtrusive chest patch sensor.
• To combine multiple physiological signals for improved respiratory rate accuracy in ambulatory settings.
• To assess the algorithm's performance across different breathing conditions and daily activities.

Main Methods:

• An algorithm was developed to combine respiratory rates derived from respiratory sinus arrhythmia (RSA), electrocardiography (ECG) QRS amplitude modulation, and tri-axial accelerometer data.
• A weighted average approach was employed, with weights determined by signal quality metrics.
• The algorithm was validated on 15 elderly subjects during metronome breathing, spontaneous breathing, and activities of daily living (ADLs).

Main Results:

• The algorithm demonstrated a low mean absolute error of 1.02 BrPM during metronome breathing.
• Errors were 1.67 BrPM during spontaneous breathing and 2.03 BrPM during ADLs when compared to a reference device.
• The combined approach effectively integrated data from multiple sensors for robust respiratory rate estimation.

Conclusions:

• The developed algorithm provides an accurate and low-complexity method for continuous respiratory rate monitoring using a chest patch sensor.
• This technology holds significant potential for enhancing remote patient monitoring and early detection of respiratory conditions.
• The algorithm's performance across various activities suggests its suitability for real-world ambulatory applications.