The non-invasive and continuous estimation of cardiac output using a photoplethysmogram and electrocardiogram during incremental exercise

Cardiac output (CO) monitoring is not only essential for critically ill patients in the hospital, but also for patients at home and those undergoing cardiopulmonary exercise testing. However, CO is difficult to monitor during daily activities and exercise. In this paper, we aim at developing a novel CO estimation method that can be used under these challenging conditions. The tube model was utilized to derive a CO index, namely the pulse time reflection ratio (PTRR) from an electrocardiogram and photoplethysmogram. After calibration, the PTRR can be used to estimate beat-to-beat CO. The proposed method was verified against CO measured by impedance cardiography on 19 healthy subjects in an incremental intensity exercise test. Results showed that there were strong correlations (r) between the PTRR and reference CO in 18 subjects (mean r: 0.88, n = 245 trials). Two calibration approaches reported in the literature were applied to the proposed method and the corresponding bias ± precisions of estimation errors were 0 ± 1.89 L min^-1 and -0.22 ± 2.12 L min ^-1, respectively. The percent errors were 21.94% and 24.90%, smaller than the clinical acceptance limit (30%). To conclude, after calibration, this method can be used to monitor CO on healthy subjects during incremental intensity exercise.