基于光电容积脉搏波的无创血压检测技术研究

发布时间：2018-05-03 09:02

本文选题：光电容积脉搏波 + 无创　；参考：《西安理工大学》2017年硕士论文

【摘要】：血压是人体的一项重要生理参数,在一定程度上反映心血管功能以及心脏和血液循环系统的变化或规律,无创、连续、实时、便捷的对血压进行监测,对高血压等心脑血管疾病的预防起着积极的作用。光电容积脉搏波(Photoplethysmography, PPG)作为人体重要的生理信号,其中包含着丰富的心血管生理病理信息,且具有操作简便、无创、低成本、安全可靠等诸多优点。本文以人体指端的光电容积脉搏波为对象,与动脉血压之间进行了一系列的分析和研究,主要研究工作有:搭建了基于STM32的人体指端光电容积脉搏波采集系统。系统选用HKG-07B型红外脉搏传感器采集人体指端容积脉搏波信号,信号经过硬件电路调理,然后通过STM32F103ZET6芯片集成的ADC进行模数转换,并通过串口把数字化的脉搏信号发送到上位机供后续处理。在上位机Matlab平台下,对采集到的经硬件调理的脉搏波信号进行去噪处理,包括毛刺和基线漂移的去除。针对脉搏波特征点检测算法进行深入研究,分析对比了微分法和小波变换法的优劣,并运用小波变换模极大值结合微分的方法准确检测出所需要的六个特征点,进一步求取到了脉搏波特征参数,与血压进行逐步回归分析,建立了人体动脉血压模型,利用该方法实现了人体血压无创连续测量。并对该方法测得结果精度进行分析,比较该方法测得值与利用水银血压计实测值之间的误差,二者结果相差基本在5mmHg范围内,符合 AAMI (Association for the Advancement of Medical Instrumentation)误差标准,进行了一致性分析,给出了 Bland-Altman分析结果,表明两种方法具有较好的一致性。
[Abstract]:Blood pressure is an important physiological parameter of human body. It reflects the changes or rules of cardiovascular function and heart and circulatory system to a certain extent. It is noninvasive, continuous, real-time and convenient to monitor blood pressure. It plays an active role in the prevention of cardiovascular and cerebrovascular diseases such as hypertension. As an important physiological signal of human body, photovolumic pulse wave (PPG) contains abundant cardiovascular physiological and pathological information, and has many advantages such as simple operation, non-invasive, low cost, safety and reliability, and so on. In this paper, a series of analysis and research are carried out between the photoelectric volume pulse wave of human finger and arterial blood pressure. The main research work is as follows: a photoelectric pulse wave acquisition system based on STM32 is set up. The system selects HKG-07B infrared pulse sensor to collect the body finger volume pulse wave signal, the signal is adjusted by the hardware circuit, and then the signal is converted by the ADC integrated with the STM32F103ZET6 chip. The digital pulse signal is sent to the host computer through serial port for subsequent processing. Based on the Matlab platform of the upper computer, the pulse wave signal which has been processed by hardware is de-noised, including burr and baseline drift removal. In this paper, the pulse wave feature point detection algorithm is deeply studied, and the advantages and disadvantages of differential method and wavelet transform method are analyzed and compared, and the six characteristic points are accurately detected by wavelet transform modulus maximum combined with differential method. Furthermore, the characteristic parameters of pulse wave were obtained, and stepwise regression analysis was carried out with blood pressure, and a blood pressure model of human artery was established. The method was used to realize the continuous measurement of human blood pressure. The accuracy of the method is analyzed, and the error between the measured value by this method and the measured value by mercury sphygmomanometer is compared. The difference between the two results is basically within the range of 5mmHg, which conforms to the standard of AAMI Association for the Advancement of Medical Instrumentation) error. The consistency analysis and Bland-Altman analysis results show that the two methods have good consistency.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R443.5;TP274

【参考文献】