An attempt of new calibration method with support vector machines regression for pulse oximetry

A novel calibration method using support vector machines regression (SVMsR) of pulse oximetry was proposed and attempted. Conventional calibration method of pulse oximetry that based on an optical density ratio of transmitted visible red light and infrared radiation whereas a proposed method here was not directly based on the optical density ratio. From theoretical point of view, conventional calibrations using the ratio can be considered as a method for nonlinear problem: nonlinear relation between two optical densities (red and IR) and oxygen saturation could be partly linearized by the ratio calculation. We thought, that nonlinear problem could be directly solved by using nonlinear analyses. Among them, the support vector machines regression method was attempted to make a calibration for pulse oximetry. As an experiment, two photo plethysmograms (PPGs) by red and IR were measured on three healthy male subjects. Simultaneously, oxygen saturation (SpO2) level was measured by a commercial pulse oxymeter. SpO2 level was controlled by breathing 10% oxygen gas for obtaining 98-76% SpO2 level. Sequentially, feature points of two PPGs were extracted in beat by beat. Convex peaks and concave valleys on PPG waveforms and moving averages of red and IR PPGs of each cardiac beat were selected as feature points. Then, nonlinear regression using SVMs were attempted to obtain relationship between SpO2 by the oxymeter (regressand) and feature points of PPG (regressor). Parameters of SVMsR were obtained by trial-and error method. In result, a regression model by trained by three-quarters of cardiac beats (teaching data) was obtained and evaluated by validation data of other one-quarter beats (validation data). As an example, 89 of 103 validation data of a subject (86% of validation data) fell within ± 1.5 % of SpO2 level by the meter. In conclusion, SVMsR could be applicable on calibration for pulse oximetry.