An automated algorithm for determining respiratory rate by photoplethysmogram in children

Abstract

Background: We have developed an automated algorithm to allow the measurement of respiratory rate directly from the photoplethysmogram (pulse oximeter waveform). Aim: To test the algorithm's ability to determine respiratory rate in children. Methods: A convenience sample of patients attending a paediatric Accident and Emergency Department was monitored using a purpose‐built pulse oximeter and the photoplethysmogram (PPG) recorded. Respiration was also recorded by an observer activating a push‐button switch in synchronization with the child's breathing. The switch marker signals were processed to derive a manual respiratory rate that was compared with the wavelet‐based oximeter respiratory rate derived from the PPG signal. Results: Photoplethysmograms were obtained from 18 children aged 18 mo to 12 y, breathing spontaneously at rates of 17 to 27 breaths per minute. There was close correspondence between the wavelet‐based oximeter respiration rate and the manual respiratory rate, with the difference between them being less than one breath per minute in all children. Conclusion: Our automated algorithm allows the accurate determination of respiratory rate from photoplethysmograms of a heterogeneous group of children. We believe that our automated wavelet‐based signal‐processing techniques could soon be easily incorporated into current pulse oximetry technology.