The Study Of Noninvasive Continuous Blood Pressure Measurement Based On Pulse Wave Transit Time

Blood pressure is an important physiological parameter that reflects the function of cardiovascular system. Compared to traditional blood pressure measurement technology, noninvasive continuous blood pressure measurement can monitor blood pressure in real time for a long period. Without shortcomings of traditional methods, noninvasive continuous blood pressure measurement technology has a considerable advantage in some special situations and in clinical. Continuous blood pressure monitoring is necessary for critically patients and surgical patients in clinical, so that when the patient in a case of an accident, health care workers can take rescue measures timely.In this thesis, two mathematical models of pulse wave transit time and blood pressure are established based on the theory of arterial blood pressure and pulse wave propagation. Different blood pressure values are obtained by changing the postural. The error between the value of blood pressure of cuff measurement and estimated blood pressure values from two mathematical models is comparative analyzed.The following research work has been carried out in this study:(1) The impact of long time acquisition and different fingertips acquisition in the data collection process is analyzed.(2) The collected signals are pre-processed. The most appropriate method for de-noising and removing baseline is determined by comparing different de-noising methods such as wavelet transform, digital filter and weighted average filter.(3) Feature parameters of ECG and pulse wave signals are extracted after pre-processed. Mexican-hat wavelet and modulus maxima method are used to detect the R wave of the ECG signals and the detection rate is above97%. The peak of the pulse wave signal is determined using the differential threshold method, and the detection rate is above98%.(4) Two models of blood pressure measurement are established, and the number of parameters is2. The parameters can be achieved by single calibration in Model-1. The two parameters in Model2are determined by the nonlinear regression of multiple measurements of blood pressure and pulse wave transit time.(5) Experiments are designed to calibrate the two models, and the models used to estimate the blood pressure compared to blood pressure measured by cuff method. The blood pressure measurement error of Model-1is less than3.5±7.15mmHg and the blood pressure measurement error of Model-2is less than2.58±6.64mmHg.