Study On Acquisition And Preprocessing Method For Photoplethysmography Signals

The Photoplethysmography(PPG)signal is an important signal of clinical monitoring equipment in hospitals,which contains a large amount of physiological information reflecting the human body,and is widely used in the evaluation of cardiovascular,respiratory and circulatory systems.However,most existing medical equipment can only calculate a few parameters such as blood oxygen saturation and blood pressure,and the signal data is not open source.It is not possible to calculate many other parameters from PPG signals,nor to analyze and process PPG signals using algorithms or models,so it is difficult to achieve diagnosis of human-related diseases and assessment of physiological status.Therefore,many researchers rely on the signal data provided by the database website for relevant research,but the quality of signals in the database is often difficult to guarantee,and and the types and quantities of the signals are different from research need,making research progress limited by existing signals.Therefore,developing a portable PPG signal acquisition system and researching preprocessing methods to improve the quality of PPG signals can provide solutions for relevant detection and analysis based on the PPG signal.This paper designs and develops a portable PPG signal acquisition system on the basis of studying the signal acquisition principle of clinical monitors,by analyzing the characteristics of PPG signal and various interferences and noises,a preprocessing strategy combining noise suppression with invalid signal elimination is proposed.Aiming at a variety of mixed noises in the signal,a PPG signal noise suppression method based on double median filter is designed to suppress the baseline drift,high frequency noise and random noise.Aiming at the invalid signal caused by poor sensor contact noise and spike noise,an invalid signal discrimination method based on the Chauvenet criterion was designed to identify and eliminate the invalid signal.Through designing the simulation and experiments,combined with test methods such as signal frequency spectrum,signal-to-noise ratio,similarity coefficient,execution efficiency and BP neural network model,the preprocessing method is tested.The results show that the preprocessing method can effectively improve the quality of PPG signal,and is conducive to improving the accuracy of relevant detection model.The PPG signal acquisition system and signal preprocessing method proposed in this paper can provide reliable data sources for the subsequent parameter calculation,analysis and processing,algorithm evaluation and relevant model construction,which have important research significance.