Blood Pressure Measurement Method And System Based On Photoplethysmography Characteristics

With the aging of the population,hypertension and other chronic diseases gradually plagued the lives of the elderly and therefore,people pay more attention to the management of personal health.However,human blood pressure values wary at every single measurement or intermittent measurements,so it is difficult to reflect the blood pressure levels.Hence,the daily intelligent wearable device and other equipment with continuous noninvasive monitoring of human blood pressure have important significance.There are some blood pressure measurement methods recently,including korotkoff auscultation method,oscillometry method,volume compensation method,arterial tension method,ultrasonic method and the pulse wave velocity method etc.All of these methods have their own advantages,but they also have some limitations in achieving noninvasive continuous measurement of blood pressure.In order to provide a feasible method for the non-invasive continuous monitoring of blood pressure in the daily health monitoring devices,the non-invasive blood pressure measurement method based on the characteristic parameters of pulse wave was investigated in the present work.In this thesis,the pulse wave signals of different human bodies were analyzed,Two new methods were proposed to extract the characteristic parameters of different types of pulse waves.The first feature parameter extraction method is based on SWT(Stationary Wavelet Transform)algorithm,the pulse wave signals were analyzed using SWT decomposition,and the fifth layer detailed coefficients were employed to reconstruct the signals.Ten characteristic parameters were obtained from the reconstructed signals.The second method was based on the EEMD(Ensemble Empirical Mode Decomposition).The original pulse wave signal was decomposed by EEMD to extract the 10 characteristic parameters of the fourth layer IMF.In this thesis,ten thousand pulse wave signals from the multi-parameter intelligent monitoring database(THE MIMIC DATABASE)were analyzed by using the two proposed methods.The correlation model of characteristic parameters and blood pressure was established by ANN(Artificial Neural Networks).The error analysis of the model was carried out.The results indicated that the error of the model meets the standards of the American Association for the advancement of medical instrumentation(AAMI).To validate the feature parameter extraction methods,a set of pulse wave signal acquisition and analysis system based on the GUIDE function of MATLAB was designed,which can collect the pulse wave signal from the human body and can estimate the corresponding blood pressure value.The system mainly includes PPG(Photoplethysmography)sensor module,NI acquisition module,signal processing and display modules.Through the analysis of the characteristics of pulse wave signals,523 nm green light was used as the light source,and the corresponding conditioning circuit was designed to collect the pulse wave signal of human fingertips efficiently.At the same time,a reasonable pulse wave acquisition program was designed to collect the pulse wave signals from different human body.The characteristic parameters of pulse wave signal were extracted by the two proposed methods.The correlation model of characteristic parameters and experimental blood pressure was established as well.Furthermore,the error test results of the model also meet the standard of AAMI.Therefore the system can be used to detect blood pressure continuously,providing some reference value for the real-time monitoring of blood pressure in the human health monitoring equipment,such as intelligent wearable devices.