OBJECTIVE: In the process of determining oxygen saturation, the pulse oximeter functions as a photoelectric plethysmograph. By analyzing how the frequency spectrum of the pulse oximeter waveform changes over time, new clinically relevant features can be extracted. METHODS: Thirty patients undergoing general anesthesia for abdominal surgery had their pulse oximeter, airway pressure and CO(2) waveforms collected (50 Hz). The pulse oximeter waveform was analyzed with a short-time Fourier transform using a moving 4096 point Hann window of 82 seconds duration. The frequency signal created by positive pressure ventilation was extracted using a peak detection algorithm in the frequency range of ventilation (0.08-0.4 Hz = 5-24 breaths/minute). The respiratory rate derived in this manner was compared to the respiratory rate as determined by CO(2) detection. RESULTS: In total, 52 hours of telemetry data were analyzed. The respiratory rate measured from the pulse oximeter waveform was found to have a 0.89 linear correlation when compared to CO(2) detection and airway pressure change. the bias was 0.03 breath/min, SD was 0.557 breath/min and the upper and lower limits of agreement were 1.145 and -1.083 breath/min respectively. The presence of motion artifact proved to be the primary cause of failure of this technique. CONCLUSION: Joint time frequency analysis of the pulse oximeter waveform can be used to determine the respiratory rate of ventilated patients and to quantify the impact of ventilation on the waveform. In addition, when applied to the pulse oximeter waveform new clinically relevant features were observed.
OBJECTIVE: In the process of determining oxygen saturation, the pulse oximeter functions as a photoelectric plethysmograph. By analyzing how the frequency spectrum of the pulse oximeter waveform changes over time, new clinically relevant features can be extracted. METHODS: Thirty patients undergoing general anesthesia for abdominal surgery had their pulse oximeter, airway pressure and CO(2) waveforms collected (50 Hz). The pulse oximeter waveform was analyzed with a short-time Fourier transform using a moving 4096 point Hann window of 82 seconds duration. The frequency signal created by positive pressure ventilation was extracted using a peak detection algorithm in the frequency range of ventilation (0.08-0.4 Hz = 5-24 breaths/minute). The respiratory rate derived in this manner was compared to the respiratory rate as determined by CO(2) detection. RESULTS: In total, 52 hours of telemetry data were analyzed. The respiratory rate measured from the pulse oximeter waveform was found to have a 0.89 linear correlation when compared to CO(2) detection and airway pressure change. the bias was 0.03 breath/min, SD was 0.557 breath/min and the upper and lower limits of agreement were 1.145 and -1.083 breath/min respectively. The presence of motion artifact proved to be the primary cause of failure of this technique. CONCLUSION: Joint time frequency analysis of the pulse oximeter waveform can be used to determine the respiratory rate of ventilated patients and to quantify the impact of ventilation on the waveform. In addition, when applied to the pulse oximeter waveform new clinically relevant features were observed.
Authors: Kirk H Shelley; Doris Tamai; Denis Jablonka; Michael Gesquiere; Robert G Stout; David G Silverman Journal: Anesth Analg Date: 2005-03 Impact factor: 5.108
Authors: Michael J Gesquiere; Aymen A Awad; David G Silverman; Robert G Stout; Denis H Jablonka; Tyler J Silverman; Kirk H Shelley Journal: J Clin Monit Comput Date: 2007-08-16 Impact factor: 2.502
Authors: Marco A F Pimentel; Alistair E W Johnson; Peter H Charlton; Drew Birrenkott; Peter J Watkinson; Lionel Tarassenko; David A Clifton Journal: IEEE Trans Biomed Eng Date: 2016-11-18 Impact factor: 4.538
Authors: Paul S Addison; James N Watson; Michael L Mestek; James P Ochs; Alberto A Uribe; Sergio D Bergese Journal: J Clin Monit Comput Date: 2014-05-06 Impact factor: 2.502