基于PPG信号的无袖带式血压估计算法研究

Research on Cuff-less Blood Pressure Estimation Algorithm Based on PPG Signal

针对传统的日常血压监测方法大都属于间断性监测,无法全面评估血压这一动态生理参数的问题,进行基于PPG信号的无袖带式血压估计算法研究。首先,从PPG信号及其导数中提取32个时域和频域特征;然后,利用Pearson相关系数对特征进行评价;接着,结合启发式搜索方法筛选出最优特征子集;最后,通过随机森林模型预测受试者的动态血压值。经实验验证,基于PPG信号的无袖带式血压估计算法满足AAMI标准,且DBP达到了BHS标准的A等级,SBP达到了BHS标准的B等级。

Aiming at the problem that traditional daily blood pressure detection methods are mostly intermittent detection,which can not comprehensively evaluate the dynamic physiological parameter of blood pressure,this paper conducts to research the cuff-less blood pressure estimation algorithm based on PPG signal.Firstly,32 features in the time and frequency domains are extracted from the PPG signal and its derivatives.Then,the Pearson correlation coefficient is used to evaluate the features,and the heuristic search method is adopted to screen out the optimal feature subset.Finally,the Random Forest model is employed to predict the blood pressure of the subject.The experimental results show that the cuff-less blood pressure estimation algorithm based on PPG signal meets the AAMI standard,while the DBP reaches grade A of the BHS standard,and the SBP reaches grade B of the BHS standard.