Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Background Our goal was to evaluate the performance of a new wearable arm located pulse oximeter. Methods Twelve volunteers were monitored with three pulse oximeters and underwent desaturation to 70% SaO(2). We compared the accuracy of SpO(2) reading from the SmartCardia system with SpO(2) using two well established devices (Masimo and Nellcor) as reference. Oximetry was performed at different level of oxygen saturation varying from 70 to 100%. Bias, ARMS and precision were evaluated using Bland-Altman plots. Results The mean (SD) differences between SaO(2) compared to SpO(2) and the devices were as follows: SaO(2) versus Masimo 2,12 +/- 1,01% (95% CI 1,45 to 2,79), SaO(2) versus Nellcor 0,78 +/- 0,58% (95% CI - 0,29 to 1,65) and SaO(2) versus SmartCardia 0,42 +/- 0,24% (95% CI - 0,64 to 1,46). The bias between SmartCardia, Masimo, Nellcor devices and SaO(2) was 0.16 (95% CI 0.05 to 0.33) and LoA (level of agreements) 2.96 (95% CI - 2,68 to 2,89) for SmartCardia, 2,02 (95% CI 1,49 to 2,54) and LoA - 6 to 11 for Masimo, and 0,76 (95% CI 0,5 to - 1) and LoA - 3,5 to 5,0 for Nellcor. ARMS for the 70-100% SaO(2) range was 1,4 for SmartCardia, 5,0 for Masimo and 2,31 for Nellcor. Conclusions The new wireless SmartCardia SpO(2) measurement system demonstrated in-line results, bias, ARMS and precision in healthy volunteers, when compared with the gold standard SaO2 and with two well established systems, Masimo and Nellcor.
Edoardo Charbon, Scott Anthony Lindner, Martin Wolf, Jingjing Jiang
Marilyne Andersen, Jan Wienold, Stephen William Wasilewski, Geraldine Cai Ting Quek