Pulse oximetry based on photoplethysmography imaging with red and green light: calibratability and challenges

Remotely measuring the arterial blood oxygen saturation (SpO 2) in visible light (Vis) involves different probing depths, which may compromise calibratibility. This paper assesses the feasibility of calibrating camera-based SpO 2 (SpO 2,cam) using red and green light. Camera-based photoplethysmographic (PPG) signals were measured at 46 healthy adults at center wavelengths of 580 nm (green), 675 nm (red), and 840 nm (near-infrared; NIR). Subjects had their faces recorded during normoxia and hypoxia and under gradual cooling. SpO 2,cam estimates in Vis were based on the normalized ratio of camera-based PPG amplitudes in red over green light (RoG). SpO 2,cam in Vis was validated against contact SpO 2 (reference) and compared with SpO 2,cam estimated using red-NIR wavelengths. An RoG-based calibration curve for SpO 2 was determined based on data with a SpO 2 range of 85-100%. We found an [Formula: see text] error of 2.9% (higher than the [Formula: see text] for SpO 2,cam in red-NIR). Additional measurements on normoxic subjects under temperature cooling (from [Formula: see text] to [Formula: see text]) evidenced a significant bias of - 1.7, CI [- 2.7, - 0.7]%. It was also noted that SpO[Formula: see text] estimated at the cheeks was significantly biased (- 3.6, CI [- 5.7, - 1.5]%) with respect to forehead estimations. Under controlled conditions, SpO[Formula: see text] can be calibrated with red and green light but the accuracy is less than that of SpO[Formula: see text] estimated in the usual red-NIR window