利用腕部脉搏波无创估测心脏每搏输出量的方法

Noninvasive estimation of left ventricular stroke volume based on pressure waves measured at the wrist

目的开发利用腕部采集的脉搏波信号无创估测心脏每搏输出量(stroke volume,SV)的算法,以实现在家庭环境中对SV的监测。方法 351名志愿者参与本研究,每名志愿者的样本数据包含桡动脉脉搏波信号和SV,前者使用小云心脉仪在腕部采集,后者利用超声设备获得,两者的测量同步进行。在剔除了39名有明显心血管异常的志愿者的数据后,样本数据被分为训练集和测试集两组,前者用于优化估测算法,后者用于检验算法的有效性。SV的估测方程基于在脉搏波信号上提取的6个特征点构建。考虑年龄、性别等生理因素对SV的影响,进一步按性别、年龄对志愿者进行二次分组,分别优化SV的估测方程。结果对两组志愿者估测的SV与超声测量值均显示良好的相关性,其中训练集的皮尔逊相关系数r=0.776,测试集的相关系数r=0.725。结论利用在腕部采集的脉搏波信号无创估测SV具有可行性,该方法有望实现在家庭环境中由使用者自主完成SV测量,对心血管健康的院外自主监测将是有益的补充。

Objective To develop a new method for noninvasively estimating left ventricular stroke volume （SV） based on pressure waves measured at the wrist so as to enable the monitoring of SV in a home- based setting. Methods There were 351 volunteers participated in the present study. For each volunteer,radial pressure waves and SV were measured by using the health cloud pulse wave instrument and an echocardiographic instrument,respectively. After excluding the data of 39 volunteers with evident cardiovascular abnormalities,the collected data were divided into the ＇training＇ group and the ＇testing＇ group so that the estimation algorithm could be optimized by using the data of the ＇ training＇ group and subsequently examined in the testing group. The SV estimation algorithm was constructed based on the six characteristic points identified on each pulse wave. In consideration of the potential influence of sex and age on SV, the estimation algorithms were optimized for subgroups of different age and sex to improve the accuracy of estimation. Results Good correlations between estimated and measured SV were observed for both groups,with the Pearson＇ s correlation coefficient being r = O. 776 for the training group and r = 0. 725 for the testing group, respectively. Conclusions Noninvasive estimation of SV from pulse waves measured at the wrist is feasible. The proposed method might enable the measurement of SV in a home-based setting, thus enriching the information available in out-of-hospital monitoring of cardiovascular health.