A Blood Pressure Measurement Method Based On Finger-tip Pulse Wave And Its Preliminary Practice On Intelligent Terminal

Blood pressure measurement based on pulse wave is an important branch of non-invasive continuous blood pressure measurement at this stage.And it can be divided into three categories including pulse wave velocity measurement method,pulse wave transit time measurement method and pulse wave parameters measurement method.Among them,the first two kinds use two channels of human physiological signals to compute blood pressure.However,the stability of the two signals affects the measurement results.The third one achieves a continuous measurement of stroke blood pressure by establishing a blood pressure measurement model using pulse wave parameters.At present,brachial artery and radial artery are commonly used in pulse wave parameters blood pressure measurement method.Blood pressure measurement models are mostly linear regression equations established using regression analysis method.And models are generally applicable to specific individual and pulse wave acquisition equipment.Similar to brachial artery pulse wave and radial artery pulse wave in shape and more easily to be measured,finger-tip pulse wave has been used for the measurement of cardiovascular physiological parameters.The rapid improvement of hardware and software of intelligent terminal makes it convenience to measure blood pressure based on fingertip pulse wave,which has a good application prospect.In this thesis,a blood pressure measurement method based on the finger-tip pulse wave is proposed after studying the relationship between finger-tip pulse wave parameters and blood pressure,and a preliminary practice was made on the mobile terminal.Firstly,the characteristics of the finger pulse wave and the existing parameters of the pulse wave were analyzed.Then some new parameters of the pulse wave are proposed for blood pressure measurement.The correlation analysis results between the parameters and blood pressure provided a theoretical basis for further study.Then,the acquisition,processing and parameters extraction methods of two kinds of pulse wave were introduced.Downloaded from the MIMIC database,the first kind of pulse wave was used to establish and evaluate the blood pressure measurement model.The second one was acquired by smart phone camera and was used to test the performance of the model in actual scene.Secondly,the establishment and evaluation of blood pressure measurement model were introduced.Partial Least Squares Regression and Back-Propagation Neural Network were used to establish the models,of which the former one can fit the relationship between the independent variables and dependents variable in the presence of high correlation between independer variables,while the later one can fit the nonlinear relationship between independent variables an dependent variables well.Evaluation results show that the Back-propagation neural networ model has better ability of blood pressure measurement.Finally,based on the background of chronic disease management on mobile terminal,a blood pressure measurement module was developed as a preliminary practice of the blood pressure measurement model,and the module was applied to an existing chronic diseases management mobile application.