Research And Implementation Of The Detection Algorithm For Human Body Physiological Index Based On Photoelectric Volume Pulse Wave

Photoplethysmography(PPG) is a non-invasive detection method used optical techniques for testing the change of blood volume in living body organization. Because blood volume will occur periodic pulse change under the action of periodic pulsation of human heart, the light intensity received by optical receiver also changes with the pulsation change, transforming optical signal gained into electrical signal, that is photoelectric volume pulse wave. Many important information of cardiovascular system is included in volume pulse wave, but so far, the physiological index obtained by photoelectric method just includes pulse rate and the degree of blood oxygen saturation. In other tests of physiological index, it just stays in the stage of theory or can not be demarcated. When human body PPG is under measurement, the response included much noise and baseline drift, because of the noises generated by electronic component, interference by electromagnetic field, the change of surround lighting and the effect of breathing and motion.Aiming at those issues above, firstly designing low-pass filter based on wavelet transform and a algorithm based on EMD(empirical mode decomposition) to remove baseline drift, the pulse of the volumetric filtering de-noising processing to wipe off high-frequency and low frequency of baseline drift with Industrial frequency. Then proposing method based on differential threshold and trapezoidal area, abstracting the characteristic parameters of PPG and use these characteristic parameters to figure out pulse rate, variability of pulse rate, arterial pressure, degree of blood oxygen saturation, stroke volume, cardiac output, peripheral resistance, arterial compliance, and so on. At the same time, this paper also proposes singularity removed algorithm based on sliding window and threshold, combining human body physiological function will exceed physiology limit of singularity removed. At last, make a relevant and regressive analysis of arterial pressure and characteristic parameters of the waveform, structuring the regression equation of systolic pressure and diastolic pressure.The experiment shows the algorithm mentioned in this thesis can effectively abstract the characteristic parameters of the photoelectric volume pulse wave, figuring out human body physiological index which conforms to the results measured by other instruments. So it can be applied to non-invasive photoelectric monitor. In addition, making use of arterial pressure and characteristic parameters of the waveform provides a new method and way for non-invasive and continuous measurement of blood pressure.