Research On Detection Of Human Physiological Parameters Based On PPG Noise Reduction Algorithm

Photoelectric volume pulse wave(Photoplethysmography,PPG)is formed by the transmission of human heart pulsation to peripheral vascular dilatation,which contains a lot of human pathological information.It can be used to measure human physiological parameters such as blood pressure,blood oxygen,heart rate and respiratory rate,and is an important source of information for the study of human health.In view of the fact that the PPG signal is easily affected by external noise,a signal denoising algorithm based on joint optimization of variational mode decomposition(Variational Mode Decomposition,VMD)and improved wavelet threshold is proposed,and a human physiological signal acquisition equipment is built.After the measured signal is denoised by the joint algorithm in this paper,the detection of blood pressure and blood oxygen saturation is realized.The main work of this paper is summarized as follows:(1)In this paper,by studying the existing denoising algorithm of photoelectric volume pulse wave signal,a joint optimized variational mode decomposition and improved wavelet threshold algorithm is designed to eliminate noise interference.The algorithm innovatively uses the difference coefficient of envelope entropy and sparrow search algorithm to optimize the parameters that need to be set for VMD:decomposition layer(k)and penalty factor(α).After decomposing the pulse wave signal with noise,multiple intrinsic mode components are obtained,and the correlation coefficients between each component and the original signal are calculated.The noise component and the effective component are judged by setting the threshold.A new improved wavelet threshold function is proposed to reduce the noise component based on the new improved wavelet threshold function.Finally,all the components are reconstructed to complete the overall denoising step.Three-axis acceleration signal screening and joint denoising algorithm are used to ensure the quality of PPG signal,which provides a good basis for signal feature point extraction.Subsequently,BP blood pressure estimation model was used to calculate blood pressure,and calibration method was used to fit the blood oxygen saturation formula to calculate blood oxygen.(2)In this paper,the front-end human physiological signal acquisition equipment is built independently,which can synchronously detect and transmit human pulse wave signal and acceleration signal,and has a certain wearability and expansibility.(3)At the end of this paper,a physiological parameter detection experiment is designed to verify the effectiveness of the proposed noise reduction algorithm through real data collection.After noise reduction,the signal-to-noise ratio(Signal-Noise Ratio,SNR)is improved to 30.2543 d B,and the root mean square error(Root Mean Square Error,RMSE)is reduced to 0.0439.The final blood pressure and blood oxygen test results are obtained by model calculation,and compared with the results of Ohlon wrist sphygmomanometer and Philips finger clip sphygmomanometer,the predicted results of blood pressure are in line with the AAMI standard and are in good agreement with omron wrist sphygmomanometer.the error of blood oxygen saturation is about 1%,and the accuracy of the overall test data meets the design requirements.Experiments show that the proposed noise reduction algorithm and human physiological parameters detection effect is good,and has a certain academic significance and application value.