基于光纤光栅脉搏波的人体血压检测方法研究

发布时间：2018-01-12 02:35

本文关键词：基于光纤光栅脉搏波的人体血压检测方法研究　出处：《天津工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着社会的老龄化、生活节奏的加快、饮食结构的变化和环境的恶化,高血压、心脏病等心脑血管疾病已成为严重威胁人类健康和生命的主要疾病。对心脑血管疾病的预防控制已经成为医学领域研究的重点之一。光纤光栅传感器体积小、灵敏度高、抗电磁干扰、与纱线兼容,在织物式传感器领域具有研究价值和应用潜力,对强电磁环境下人体生理参数的监测具有重要意义。本文在课题组前期研制的光纤光栅脉搏织物传感器的基础上,对基于光纤光栅脉搏波的人体血压检测方法进行了研究。研究内容主要分为两部分:光纤光栅脉搏波信号处理算法和基于多脉搏特征参数的人体血压估算模型的建立方法。结合光纤光栅脉搏波的特点,提出了一种小波阈值消噪与改进数学形态学滤波相结合的光纤光栅脉搏波消噪算法,并根据脉搏周期对形态学滤波中结构元素的长度进行自适应选择,从而改善了去除基线漂移的效果。研究了脉搏波特征点的生理意义,设计了一种脉搏波特征点的提取算法,提高了脉搏波峰值点和起止点检测的准确性。实验结果表明,经消噪处理后,光纤光栅脉搏波的信噪比提高了一倍。在误差允许的范围内,脉搏波峰值点和起止点提取的准确率分别达到了 97.2%和97.6%。搭建了人体多生理参数采集实验平台,可对光纤光栅脉搏波、压电式脉搏波以及光电容积脉搏波进行同步采集,并基于LabVIEW编写了信号处理软件,实现了光纤光栅脉搏波的预处理及相关脉搏特征的提取。在对脉搏特征与人体血压相关性研究的基础上,提出了一种基于多脉搏特征的个性化非线性人体血压估算模型的建立方法。首先选取被测个体与血压相关性较高的脉搏特征作为建模参数;然后,分析这些脉搏特征与被测者血压的函数关系;最后,基于上述脉搏特征及其与血压的函数关系通过最小二乘拟合算法建立针对该个体的脉搏-血压估算模型,实现收缩压与舒张压的无袖带式连续测量。实验结果表明,本文提出的个性化非线性血压估算模型对于收缩压和舒张压的平均计算误差分别为1.93±1.39mmHg和2.05±1.56mmHg,满足血压测量国际标准对于无创血压测量精度的要求。在此基础上,将被测者分为四类,提出了针对每一类人群的通用血压模型建立方法并进行了初步研究。实验表明,该模型对收缩压和舒张压的计算误差分别为3.56±2.68mmHg和4.88±3.76mmHg。通过增加每类人群参与建模的样本数量,可使通用血压估算模型的计算精度进一步提高。本文的研究成果将为基于光纤光栅传感器的人体生理参数检测方法的研究提供参考和借鉴,也为可穿戴无袖带连续血压测量技术的发展提供重要依据。
[Abstract]:With the aging of society, the accelerated pace of life, changes in dietary structure and deterioration of the environment, hypertension. Cardio-cerebrovascular diseases such as heart disease have become a serious threat to human health and life. The prevention and control of cardiovascular and cerebrovascular diseases has become one of the focuses in the field of medicine. Fiber Bragg grating sensors are small in size. High sensitivity, anti-electromagnetic interference, compatible with yarn, it has research value and application potential in the field of fabric sensor. It is of great significance to monitor the physiological parameters of human body under strong electromagnetic environment. This paper is based on the fiber Bragg grating pulse fabric sensor developed by our research group. The method of blood pressure measurement based on fiber Bragg grating pulse wave is studied. The research is divided into two parts:. The signal processing algorithm of fiber Bragg grating pulse wave and the method of establishing blood pressure estimation model based on multi-pulse characteristic parameters. A pulse wave de-noising algorithm of fiber Bragg grating (FBG) is proposed, which combines wavelet threshold de-noising with improved mathematical morphological filtering, and adaptively selects the length of structural elements in morphological filtering according to pulse period. In order to improve the effect of removing baseline drift, the physiological significance of pulse wave feature point is studied, and a pulse wave feature point extraction algorithm is designed. The experimental results show that the SNR of fiber Bragg grating pulse wave is doubled after denoising. The accuracy of pulse peak extraction and starting and ending point extraction are 97.2% and 97.6.The experimental platform of human body multi-physiological parameters acquisition can be used for fiber Bragg grating pulse wave acquisition. The piezoelectric pulse wave and the photoelectric volume pulse wave are collected synchronously, and the signal processing software is programmed based on LabVIEW. The pretreatment of fiber Bragg grating pulse wave and the extraction of related pulse characteristics are realized. Based on the study of the correlation between pulse characteristics and human blood pressure. In this paper, a method of establishing a personalized nonlinear human blood pressure estimation model based on multi-pulse features is proposed. Firstly, the pulse characteristics which are highly correlated with blood pressure are selected as the modeling parameters. Then, the functional relationship between these pulse characteristics and the blood pressure of the subjects was analyzed. Finally, based on the above characteristics of pulse and its relationship with blood pressure, the estimation model of pulse and blood pressure for the individual is established by least square fitting algorithm. The sleeveless continuous measurement of systolic pressure and diastolic pressure is realized. The calculated errors of the personalized nonlinear blood pressure estimation model for systolic and diastolic blood pressure were 1.93 卤1.39mmHg and 2.05 卤1.56mmHg, respectively. To meet the requirements of the international standard of blood pressure measurement for non-invasive blood pressure measurement accuracy. On this basis, the subjects are divided into four categories. In this paper, a general blood pressure model for each population is proposed and a preliminary study is carried out. The calculation errors for systolic and diastolic blood pressure were 3.56 卤2.68 mmHg and 4.88 卤3.76 mmHg, respectively. The research results of this paper will provide reference and reference for the research of human physiological parameters detection method based on fiber Bragg grating sensor. It also provides an important basis for the development of wearable sleeveless continuous blood pressure measurement technology.
【学位授予单位】：天津工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R443.5;TP212

【相似文献】