Extraction of respiratory rate from PPG using ensemble empirical mode decomposition with Kalman filter

This Letter suggests a simple but effective approach for accurate estimation of respiratory rate (RR) from the photoplethysmogram (PPG). In the suggested technique, the PPG signal is first decomposed into a number of intrinsic mode functions (IMFs) using the ensemble empirical mode decomposition (EEMD). The IMFs comprising respiratory information are selected and used to reconstruct the signal. A Kalman filter coupled with the signal quality measure of PPG is utilised to process the reconstructed signal to derive the respiratory signal for RR estimation. Experiments are conducted on two independent datasets, namely CapnoBase and MIMIC. Based on experimental results, the technique is found to be outperforming the existing methods by achieving lower error in the estimated RRs on both datasets. This work demonstrates the potential of the EEMD method with Kalman filter for precise estimation of RR from PPG to contribute to the advancement of portable healthcare systems with a fewer number of sensors.