Respiratory rate estimation using respiratory sinus arrhythmia from photoplethysmography

Area of Science:

• Biomedical Engineering
• Physiological Monitoring
• Signal Processing

Background:

• Respiratory rate (RR) is a critical vital sign in healthcare.
• Unobtrusive, mobile monitoring of RR is highly desirable for ambulatory care.
• Photoplethysmography (PPG) signals from pulse oximeters are readily available but often have suboptimal sampling rates.

Purpose of the Study:

• To investigate the feasibility of estimating RR from sub-optimally sampled PPG signals.
• To explore the extraction of respiratory sinus arrhythmia (RSA) from PPG-derived heart rate variability (HRV) for RR estimation.
• To compare the performance of PPG-based RR estimation against electrocardiography (ECG).

Main Methods:

• Analysis of PPG and ECG signals from 29 children and 13 adults under general anesthesia.
• Extraction of RSA from PPG-derived HRV and comparison with ECG-derived RSA.
• Reference RR was determined using capnography.
• Real-time algorithms were employed for RSA extraction and RR estimation.

Main Results:

• PPG-derived RSA power was significantly higher than ECG-derived RSA power (182.42 ± 36.75 dB vs. 162.30 ± 43.66 dB).
• Mean RR error was lower for PPG compared to ECG.
• Both PPG and ECG methods showed higher reliability during spontaneous ventilation versus pressure-controlled ventilation.
• PPG signal artifacts significantly increased RR error, a trend less pronounced with controlled ventilation.

Conclusions:

• Estimation of RR from sub-optimally sampled PPG signals is feasible.
• PPG-based RR estimation is more reliable and powerful than ECG-based estimation.
• Ventilation mode and signal artifacts impact the accuracy of PPG-based RR monitoring.