Automated Hilbert Envelope Based Respiration Rate Measurement from PPG Signal for Wearable Vital Signs Monitoring Devices

Respiratory rate (RR) is one of the most vital signs to predict symptoms of serious illnesses and also used as a vital indicator or significant physiological parameter for early disease warning (early detection of patient deterioration) and to monitor person’s physical and emotional stress. In this paper, we propose an automated Hilbert envelope based respiration rate estimation method using the photoplethysmogram (PPG) signal. The proposed Hilbert transform RR (HT-RR) method is tested by using the signals taken from BIDMC and CapnoBase databases. On the benchmark performance metrics, the proposed method had an mean absolute error (MAE) in terms of median (25th–75th percentile) of 3.7(1.8–5.5) breaths per minute (brpm) and 2.6 (0.8–5.5) brpm for 30 and 60 second PPG signals respectively. Evaluation results further showed that the processing time of 4.81 ± 0.80 milliseconds are required to compute RR value from 30 seconds duration PPG signal. The method has great potential in improving the accuracy and reliability of wearable and portable diagnosis system. It is observed that the proposed method outperforms the recent RR estimation methods.