A cost-effective and video-based method for continuous cardiac output monitoring: design and clinical validation

Background Continuous monitoring of cardiac output (CO) is limited by expensive consumables and equipment, which is rarely used in resource-limited settings. The aim of this study was to develop a novel method based on video analysis technique (called “video-based method”) to estimate CO deriving from ubiquitous smartphones. Methods Videos of minimal invasive radial arterial waveform were recorded and analyzed by a smartphone. CO values obtained from the FloTrac/Vigileo system, and video-based method were collected simultaneously. Further, we compared the performance of video-based method with commercial FloTrac/Vigileo system for clinical validation. Concordance and interchangeability of CO measurements were assessed using the intra-class correlation coefficient (ICC), mean error and the Bland-Altman analysis (B-A). Furthermore, the ability of detecting directional changes in CO values was also evaluated. Results In the clinical validation section, eleven patients who received general anesthesia were included, with a total amount of 3,892 pairs of CO values acquired. In B-A, the average CO value extracted by video-based method was 0.11 (95%CI, 0.10–0.14; p < 0.001) L/min, which was lower than that provided by the FloTrac/Vigileo system. Strong concordance (ICC = 0.87; p < 0.001) and acceptable interchangeability (mean error = 27.65%) were achieved, which indicated excellent clinical validity. Additionally, the sensitivity for significant directional changes was 92%, and the specificity was 62%. Conclusions A video-based method for cost-effective and continuous CO monitoring was successfully developed, which evaluated continuous CO values by analyzing arterial pressure waveform of recording monitor videos in real time. This video-based method provided extraordinary clinical performance, thus offering a cost-effective alternative for real-time CO monitoring.