Study Of The Continuous Measurement Of Blood Pressure Based On The Characteristics Of Photoplethysmography

Blood Pressure(BP)is an important indicator for assessing cardiovascular health in the human body.Studies have shown that continuous BP monitoring has important clinical significance for the monitoring of acute critical illness and chronic disease prognosis management.In recent years,the cuffless technology of the measurement of BP has received extensive attention.And the PTT-based cuffless measurement of BP is considered to be the most promising wearable technology,however,the accuracy of this method for patients with hypertension remains to be solved.Therefore,it is important to study the continuous BP measurement algorithm based on PTT.Studies show that PPG contains rich physiological information,however,the traditional PTT-based BP estimation algorithm is limited to the acquisition accuracy of PTT,and ignores the research on the waveform characteristics of PPG.Based on the mechanisms of the formation of BP and PPG,this paper extracts the physiological parameters that characterize BP changes from PPG and studies the physiological relationship between the waveform of PPG and the vasoconstriction and expansion motion of blood vessel.And based on the generation of BP and the long-term and short-term adjustment mechanism,a new BP estimation algorithm based on PPG is established.Based on the profound study of the waveform of PPG,this paper extracts the main pulse ampltitude h as the physiological parameter to characterize the change of blood vessel and the resistance of peripheral arterial.By comparing the spectrum distribution of h of different populations,it is found that the spectrum of the healthy people’s h distributes uniformly in the low frequency band and the high frequency band,which reflects the balanced regulation of BP by sympathetic and parasympathetic nerves.However,the spectrum of the hypertensive patients’ h moves to the low frequency band,which is consists with the spectrum of the HRV.The non-uniform distribution of the spectrum of h and HRV indicates that the sympathetic nerves are more sensitive than parasympathetic nerves in the hypertensive patients.Therefore,h can be used to track the change of BP in the low frequency band.At the same time,the Multi-Gaussian model is applied to fit the PPG signals to decompose the forward incident wave and the first reflection wave,and calculate the parameter AIx to reflects the stiffness of vascular.To improve the accuracy of the estimation of BP,we combine the physiological parameter h and AIx used as a compensation factor to establish a new blood pressure estimation algorithm.Mean and standard deviation(SD)of the difference between estimated BP with the proposed model and reference BP to evaluate the accuracy.The proposed method was validated in 124 subjects and the estimated SBP and DBP with proposed method in healthy people were 0.48 ± 5.49 mm Hg and-0.43 ± 5.31 mm Hg respectively,while the estimated BP with proposed method in hypertensive patients were 0.81 ± 7.14 mm Hg and 1.49 ± 7.45 mm Hg respectively.Comparing with the PTT-based method,especially for the hypertensive patients,there is a great improve in the accuracy of the estimated BP values,which suggests that the performance of the proposed method is better than PTT-based method and meets the standard of AAMI.