Evaluation of arterial oxygen saturation using RGB camera-based remote photoplethysmography

摘要

Plethysmogram is the periodic variation in blood volume due to the cardiac pulse traveling through the body. Photo-plethysmograph (PPG) has been widely used to assess the cardiovascular system such as heart rate, blood pressure, cardiac output, vascular compliance. We have previously proposed a non-contact PPG imaging method using a digital red-green-blue camera. In the method, the Monte Carlo simulation for light transport is used to specify a relationship among the RGB-values and the concentrations of oxygenated hemoglobin (C_(HbO)) and deoxygenated hemoglobin (C_(HbR)). The total hemoglobin concentration (C_(HbT)) can be calculated as a sum of C_(HbO) and C_(HbR). Applying the fast Fourier transform (FFT) band pass filters to each pixel of the sequential images for C_(HbT) along the time line, two-dimentional plethysmogram can be reconstructed. In this study, we further extend the method to imaging the arterial oxygen saturation (SaO_2). The PPG signals for both C_(HbO) and C_(HbR) are extracted by the FFT band pass filter and the pulse wave amplitudes (PWAs) of C_(HbO) and C_(HbR) are calculated. We assume that the PWA for C_(HbO) and that for C_(HbR) are decreased and increased as SaO_2 is decreased. The ratio of PWA for C_(HbO) and that for C_(HbR) are associated to the reference value of SaO_2 measured by a commercially available pulse oximeter, which provide an empirical formula to estimate SaO_2 from the PPG signal at each pixel of RGB image. In vivo animal experiments with rats during varying the fraction of inspired oxygen (FiO_2) demonstrated the feasibility of the proposed method.