脉图信息采集及分析方法研究

【作者】 白金星；

【导师】 赵子婴；

【作者基本信息】 山东大学 ， 信号与信息处理， 2006， 硕士

【摘要】 脉搏是常见的生理现象，是心脏和血管状态等重要信息的外在反映。中医脉诊经过2600多年的实践，积累了丰富的临床经验，已经发展成为一门具有卓越和独特诊断技巧的“中医脉学”，在国际上越来越受到重视。然而由于种种原因，中医脉象的分辨标准仍停留在主观上，因此，中医脉诊的客观化成为现代中医学研究的一个重要方面。本论文研究了基于微机的脉搏信号采集、处理、分析系统，研究工作的主要结果和创新点可归纳如下： (1)研究了脉搏信号采集电路的设计。分析研究了PVDF压电薄膜的压电方程，设计出了便于中医取脉施压的采集电路的敏感部分；根据人体脉搏信号的特点，设计出了脉搏信号采集电路，并对采集中可能引入的噪声进行了分析，根据噪声的性质采取不同的措施加以滤除。 (2)对脉搏信号小波变换滤波进行了较深入的研究，应用小波变换模极大值滤波方法对硬件没有滤除的噪声进行了进一步的滤除，取得了很好的滤波效果。利用小波变换提取了脉搏信号的特征值，分析了正常的人脉搏信号的小波域特征，发现脉搏信号的时域特征点严格对应于其小波变换的过零点。提出的过零点检测脉搏信号时域特征点的算法，较好解决了脉搏信号时域特征点参数多，分析复杂，误差较大的问题。并对正常人脉搏信号和妊娠脉的特征进行分析，为临床应用提出了一个简易的判断依据。 (3)分析了妊娠脉搏的特性。通过对45例妊娠脉的分析发现妊娠高血压综合征脉搏与健康妊娠脉的明显区别是妊高征者脉搏信号缺少TWBH波，而这一症状正是由于妊高征者小动脉血管痉挛所引起的。由此分析得出了妊高征脉搏信号的变异性，为妊高征的早期预测提供了一种新方法。 (4)为了便于对采集的脉搏信号进行分析，本文设计了基于凌阳单片机的A／D和232串口通信并与VB编程相结合的脉搏分析系统，实现了脉搏信号的实时采集、波形显示、数据存贮等功能，为脉搏分析客观化提供了数据基础。更多还原

【Abstract】 Pulse is a common physiological phenomenon, it can depict the heart and blood vessel state through its wave. Diagnosis by feeling the pulse of Chinese medicine through more than 2600 year practice, it has been accumulated abundant clinical experience, and it had developed into a kinds of the remarkable unique to diagnose skill, it is " Chinese medicine diagnosis by feeling the pulse" and it has been paid attention to in the world. For various reasons, the resolution standard of the diagnosis by feeling of Chinese medicine still stays subjectively, so that the studying that the diagnosis by feeling the pulse of Chinese objectifying an important respect becoming modern traditional Chinese medicine .This thesis has studied base on the computer gather the pulse signal of the real time, deals with and it’s analysis system. The main result and innovation of research work in this paper can be summed up as follows:The first that we studied the design of the circuit about the pulse signal collecting. In this parts we analysis and research PVDF piezoelectricity membrane and it’s equation of piezoelectricity, we designed to the parts of sensitive for the circuit that it can very easy to imitate Chinese medicine fetch the pulse. According to the characteristic of the human pulse signal, we designed to the circuit for gathering to the pulse signal. And we analyzed which the noise to be lead into gather and get rid of it using corresponding measure based on different noises.The second that we deeply researched the wavelet transform filtering for pulse signal. We farther filtered the noise that the hardware didn’t filter to use the mould maximum filtering method and have get well result. We have get of the characteristic value of the pulse signal using the wavelet transform. We analyzed the pulse signal of the normal person’s wavelet domain characteristic and find that between the pulse time domain characteristic and the wavelet transform zero-crossing-was that the absolute corresponding. So we advanced the algorithm that analyzed the pulse signal used the wavelet transform zero-crossing. This method better solved that the pulse signal have a lots of characteristic parameter in time domain and it was complicated when we analyzed it and being biggish error. Go on, we analyzed normal people pulse signal and更多还原