运用小波模极大值滤波算法消除光电容积脉搏波中的运动干扰  

Motion artifact cancellation in photoplethysmography using reconstruction of wavelet transform modulus maxima

作　　者：张坤[1] 焦腾[1] 付峰[1] 张雯[1] 董秀珍[1] 

机构地区：[1]第四军医大学生物医学工程系医学电子工程教研室,西安710032

出　　处：《仪器仪表学报》2009年第3期586-589,共4页Chinese Journal of Scientific Instrument

基　　金：'十一五'军事医学专项课题(06Z046)资助项目

摘　　要：在通过光电容积脉搏波(PPG)测量血氧饱和度的过程中,PPG信号易受到被测者身体移动的影响产生运动伪迹,这些无规律的运动伪迹影响了血氧饱和度测量的稳定性和准确性。首先利用朗伯-比尔(Lamber-Beer)定理和指尖半球形模型,分析了运动干扰与正常信号之间的关系,分析表明运动干扰信号与正常信号约成线性叠加关系。并利用分析结果引入小波模极大值重构滤波算法去除信号中的运动干扰,而后通过初步实验将该算法与常规快速傅里叶(FFT)算法进行了比较。实验结果表明,此方法较FFT法有更高的准确性和稳定性。Body movement during measurement can generate motion artifacts in photoplethysmography （PPG）. The motion artifacts reduce the accuracy of PPG signal analyses. This paper first analyzes the relationship between the motion artifact and PPG signals using the Beer- Lambert law and the fingertip hemispheric model. The analytic result shows that the corrupted PPG AC signals are the sum of the normal PPG AC signals and the mutative signals caused by the movement of the tester. Based on the results, we try to use the reconstruction of wavelet transform modulus maxima to cancel the motion artifact in the corrupted PPG signals. Preliminary comparison experiments show that the wavelet method is more stable and accurate than FFT method.

关 键 词：小波 模极大值 光电容积脉搏波 运动伪迹 

分 类 号：R318.04[医药卫生—生物医学工程]