Objective: We studied the application of our algorithm for the robust extraction of respiratory information from the pulse oximeter signal acquired from a selection of patients attending the chest clinic.
Methods: Photoplethysmograms were obtained from 16 individuals: 13 patients with various conditions in the respiratory ward and three healthy subjects. Wavelet transforms were generated from which respiratory information was extracted to obtain a measure of respiratory rate. This measured rate was compared with the respiratory rate determined by one of a variety of other means (a digital end tidal CO(2) signal, the output from a non-invasive ventilation device, or a switch actuated by the patient or observer.)
Results: Respiratory rates varied from 6.2 to 35.8 breaths per minute (bpm). The oximeter rate determined through our method matched the marker rate obtained for all patients to within 1 bpm.
Conclusion: The technique allows the measurement of respiratory rate directly from the photoplethysmogram of a pulse oximeter, and leads the way for development of a simple non-invasive combined respiration and saturation monitor useful for patients with all forms of breathlessness.