A model-based artifact reduction method for the non-invasive estimation of blood oxygen saturation

We present a novel motion-artifact (MA) insensitive method for the estimation of blood oxygen saturation (SpO2), based on a heuristic physical model of the photoplethysmographic(PPG) signal with a multiplicative component representing motion artifact. This new method calculates the ratio of ratios R using the first derivative values of zero-crossing points of PPG signals, which possesses the inherent insensitivity to motion artifact. Four types of motions, as the source of motion artifact, were investigated to test the performance of the new approach. A quantitative comparison between the conventional method and the newly proposed method was undertaken based on the practical experiments. It was found that the new approach had a significant reduction in the estimation error of SpO2, compared with the conventional method