摘要
据报道,血压的变化和脉搏波形息息相关,可以利用脉搏特征参数间接估计血压。本文针对无创血压的连续监测提出一种新的预测模型。本研究从脉搏波信号波形图中提取14个特征参数,建立多元线性模型,通过改进的逐步回归算法动态实时估计连续血压,并利用多参数智能监控数据库中6个病人的数据来验证该动态模型。分析实验结果发现,血压估计偏差的均值和方差符合美国医疗器械AAMI标准(5±8 mm Hg),且预测的血压值和提取的血压值呈显著相关。实验表明,利用脉搏波这14个特征参数参数建立的动态模型可以被用来预测连续血压,而且预测正确率较高。
Pulse wave is reported to have a correlation with blood pressure and character parameters of pulse wave,which canpredict blood pressure indirectly.A new predictive model was proposed in this paper for continuous monitoring of noninvasive bloodpressure.This study aimed to extract14parameters from the pulse wave signal in order to establish the multivariate linear model.The improved stepwise regression analysis algorithm was used to estimate the blood pressure in real time,and the data of six patientsin the MIMIC database was used to validate the dynamic model.The experimental results indicated that the mean and variance ofdeviation of blood pressure estimation met the American Medical Device AAMI standard(5±8mmHg),and the predicted bloodpressure was significantly correlated with the extracted blood pressure values.The results show that the dynamic model establishedby14parameters of pulse wave can be used to predict continuous blood pressure with high accuracy.
作者
谢寒霜
王瑞平
王艳洁
刘于豪
XIE Hanshuang;WANG Ruiping;WANG Yanjie;LIU Yuhao(Department of Biomedical Engineering, School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China)
出处
《中国医疗设备》
2017年第10期39-43,共5页
China Medical Devices
关键词
脉搏波特征参数
无创血压
逐步回归算法
动态模型
character parameters of pulse wave
non-invasive blood pressure
stepwise regression analysis algorithm
dynamic model
