Study Of Photoplethysmography Based Pulse Wave Modeling And The Anti-motion Technology In Its Detection

Pulse wave signal is one of the most important physiological parameters of human body,which not only directly indicates the heart rate,but also is closely connected to other indicators of human body such as blood pressure and oxygen saturation.Moreover,pulse can reflect many physiological and pathological features of respiratory system,cardiovascular system,etc.Photoplethysmography(PPG)is a method that measures the blood volume of blood vessels via electro-optic technique.As a kind of non-electrophysiological signal with low frequency,PPG signal cannot be easily distinguished for motion-artifact overlapping.Motion-artifact reduction has been a tough problem in the area of physiological signal detection.Firstly,aimed on the source of motion-artifact,a new reflection model for motion artifact reduction based on electro-optic technique is proposed.The model is designed based on a dual-wavelength reflective PPG detection model.A high-reflection film with wavelength selectivity is used.One light channel can totally pass through the film to detect the mixed signal of motion-artifact and real pulse signal.Another light channel is completely reflected on the upper surface of the film layer to extract pure motion-artifact as a reference signal for subsequent signal processing,thus improving the system’s anti-motion interference capability.Then,based on the Monte Carlo method,a MATLAB simulation program is written to analyze the transmission of light in human skin.The relationship between pressure pulse wave waveform and human skin optical parameters is further established by pressure pulse wave transmission COMSOL finite element analysis and dynamic optical human skin modeling.Furthermore,Monte Carlo simulation and dynamic optical model are combined to propose a dynamic PPG modeling method.Finally,the paper verifies the effectiveness of the motion artifact reduction via both simulation and experiment.On the one hand,using the dynamic PPG modeling method proposed in this paper,the model is simulated,and several kinds of motions(medium frequency motion,high frequency motion and jitter)are simulated.The methods of adaptive filtering and independent component analysis are used for signal processing.The anti-motion interference ability of the model is verified from calculating the correlation with the real signal,waveform shape analysis and spectrum analysis.On the other hand,a reflective PPG acquisition device with miniaturization,low power consumption and Bluetooth communication function was developed as an experimental platform to build the motion-artifact reduction model.Several kinds of motions(low frequency motion,medium frequency motion and high frequency motion)are also simulated.1800 data segments were acquired for analysis,which verifies the validity of the model.The anti-motion model based on optical method proposed by the research improves the existing problems from a new perspective,which has high application value.The dynamic PPG modeling method based on Monte Carlo method proposed in this paper also provides a powerful tool for subsequent research.