小波函数在内源光信号去噪中的应用  

Application of Wavelet Function in Intrinsic Optical Signals De-noising

作　　者：侯钧杰 邵双运[1] 郭学谦[2] 何玉环 彭燔 王萍 HOU Junjie;SHAO Shuangyun;GUO Xueqian;HE Yuhuan;PENG Fan;WANG Ping(School of Science,Beijing Jiaotong University,Beijing 100044,China;School of Biomedical Engineering(Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application),Capital Medical University,Beijing 100069,China)

机构地区：[1]北京交通大学理学院,北京100044 [2]首都医科大学生物医学工程学院(首都医科大学临床生物力学应用研究北京市重点实验室),北京100069

出　　处：《中国医疗设备》2020年第2期81-85,89,共6页China Medical Devices

基　　金：北京市教育委员会科技计划一般项目（KM201610025010）

摘　　要：目的内源光信号(Intrinsic Optical Signals,IOSs)探测技术因受到低信噪比的影响,阻碍其对视网膜功能的非侵入性检查的实际应用。本文利用小波函数对IOSs进行去噪。方法除了稳定的探测系统外,另一个重要的方法是利用去噪算法对信号进行处理来提高信噪比。本文对比多组数据叠加取平均、小波去噪以及平滑去噪的结果最终选取了Symlets小波函数对IOSs进行去噪处理。结果实验结果表明,采用Symlets小波函数对IOSs去噪,在两组实验数据平均的情况下,信噪比得到了有效提高,较5组数据(每组实验数据采集时间为8 s,连续采集5组)平均的信噪比最高可提高3.56倍。Symlets小波函数对信号进行处理时所需要采集数据的时间,较5组数据平均所需的时间缩短24 s,信号处理所需时间平均缩短75 s左右。结论利用小波函数处理信号,在数据采集过程中可以有效降低重复采样组数,有利于IOSs的实际应用。本文结果对于视网膜IOSs的分析研究具有一定的意义。Objective Intrinsic optical signals(IOSs)detection technology is affected by low signal-to-noise ratio(SNR),which hinders its practical application in non-invasive examination of retinal function.In this paper,the wavelet function was used to denoise the IOSs.Methods In addition to a stable detection system,another important method is to use a de-noising algorithm to process the signal to improve the SNR.In this paper,the Symlets wavelet function was selected to de-noise the IOSs by comparing the results of averaging multiple groups of data,wavelet and smoothing de-noising.Results The experimental results showed that the SNR of the IOSs was effectively improved by using Symlets wavelet function when the data were averaged over two consecutive trials,which was up to 3.56 times higher than the SNR of the data that were averaged over five consecutive trials(each group of experimental data acquisition time was 8 s,and 5 groups were collected continuously).When the Symlets wavelet function processed the signal,the time required to collect the data was shortened by 24 s compared to the average time of five consecutive trials,and the time required for signal processing was reduced by about 75 s on average.Conclusion Using wavelet function to process signals can effectively reduce the number of repeated sampling in the process of data acquisition,which is beneficial to the practical application of IOSs.This study has a certain significance for the analysis and research of retinal IOSs.

关 键 词：小波函数 信号去噪 图像处理 内源光信号 

分 类 号：R318[医药卫生—生物医学工程]