| 基于高斯拟合的神经网络血压测量算法 |
| |
| 引用本文： | 温亮,李振波陈佳品等.基于高斯拟合的神经网络血压测量算法[J].传感器与微系统,2014(4):132-134,138. |
| |
| 作者姓名： | 温亮 李振波陈佳品等 |
| |
| 作者单位： | 上海交通大学微纳科学技术研究院薄膜和微细技术教育部重点实验室,上海200240 |
| |
| 基金项目： | 国家自然科学基金资助项目（61175100,51275285）;上海交通大学医工交叉基金资助项目（YG2011ZD01）;中国博士后基金资助项目（2012M510087） |
| |
| 摘 要： | 针对示波法血压测量的准确性提出一种改进方法。和以往脉搏波的提取方法不同,采用2个高斯函数之和作为模型对脉搏波包络线进行拟合。以拟合后的高斯函数参数为输入,通过2个前馈神经网络进行收缩压和舒张压的判定。实验表明:和幅度系数法相比,该算法在准确性上有了明显提高。
|
| 关 键 词： | 血压测量 高斯拟合 神经网络 示波法 |
Blood pressure measurement algorithm of neural network based on Gaussian fitting |
| |
| WEN Liang,LI Zhen-bo,CHEN Jia-pin,ZHANG Da-wei.Blood pressure measurement algorithm of neural network based on Gaussian fitting[J].Transducer and Microsystem Technology,2014(4):132-134,138. |
| |
| Authors: | WEN Liang LI Zhen-bo CHEN Jia-pin ZHANG Da-wei |
| |
| Affiliation: | 1.Key Laboratory for Thin Film and Microfabrication of Ministry of Education, Research Institute of Micro/Nano Science and Technology, Shanghai Jiaotong University, Shanghai 200240, China;) |
| |
| Abstract: | An improved method is presented to increase accuracy of oscillometric blood pressure measurement.Unlike previous extraction methods,the oscillometric waveform envelope（OMWE） is mathematically modeled as a sum of two Gaussian functions.Taking parameters of fitted Gaussian function as input,through two feed-forward NNs to determine systolic and diastolic blood pressure.Experiments prove that the accuracy of this algorithm is effectively improved compared with amplitude coefficient method. |
| |
| Keywords: | blood pressure measurement Gaussian fitting neural network oscillography |
| 本文献已被 CNKI 维普 等数据库收录！ |
