一种新的人体呼吸波采集方法与实现

A New Human Respiratory Wave Acquisition Method and Implementation

目的:人体脉搏信号中包含有丰富的生理信息,如血氧、心率、呼吸、血压等。利用光电容积描记法获得的脉搏波被称为光电容积脉搏波(Photolethysmography,PPG),并因无创、多参数测量、操作简单、成本低等特点,日益被生物医学工作者重视和青睐,因此高质量脉搏信号的获取及有效的处理方法对临床诊断分析具有重大意义。为了临床上获取呼吸波的简便易行,依据现有的脉搏信号检测技术和常用处理方法,搭建了一种新型的硬件电路,实现光电容积脉搏波的无创采集,进而提取呼吸信号。方法:通过研究人体指端光电容积脉搏波的物理特性,采用经验模态分解算法对系统获得的脉搏波信号进行模态分解,选择具有适当频率的本征模函数重构出待测的呼吸波信号,实现临床所需的人体呼吸波信号采集。利用现有的临床设备同步采集人体鼻端呼吸信号,作为标准参考呼吸信号。结果:对比两种方法获得的呼吸信号,进行时频域的分析及计算相关参数,包括个数、周期SD(Standard Deviation)值、峰值周期和AR(Auto Regressive)功率谱等,可知两种呼吸波波形具有很好的相关性,相关系数达到了0.8左右;AR功率谱相关系数达到0.8以上。结论:应用经验模态分解算法可以有效地提取光电容积脉搏波中的呼吸波成分,这就为临床呼吸波的采集与分析提供了一种新的方法。

Objective：The pulse signal of human body contains abundant physiological information,such as oxygen,heart rate,respiration,blood pressure and so on.The pulse wave obtained by the method of NIRP was called photoplethysmography,which had attracted more and more attention and favor by the biomedical researchers for the advantages of noninvasive,variety of parameters measurement,simple operation,low cost.Therefore,high quality signal acquisition and processing method effectively is of great significance to the clinical diagnosis analysis.In order to obtain the clinical respiratory wave conveniently,a new software and hardware system was introduced to achieve noninvasive acquisition of photoplethysmography（PPG） signals making use of the breath wave noninvasive detection technology.Methods：Studying physical characteristic of the human fingertip PPG signals,the pulse wave signals were decomposed using the method of empirical mode decomposition（EMD） to recombinate the respiratory signals which were needed in clinical detection.At the same time to collect human nasal respiratory signal synchronization using clinical existing equipment as the standard reference respiratory signal.Results：By doing some time frequency analysis and calculation ofrelated parameters,including the number,cycle SD（Standard Deviation） value,peak period and AR（Auto Regressive） power spectrum,we can know that the two respiratory signals had good correlation,the waves correlation coefficient was0.8 andthe AR power spectrum correlation coefficient reached0.8.Conclusions：The hardware circuit can get extract a good human finger tip PPG signals and empirical mode decomposition method can effectively extract the respiratory components,which provide a new method for collection and analysis of clinical respiratory wave.