查询结果:   朱贝贝,赵菊敏,李灯熬.基于吉文斯变换提取ECG中J波信号[J].计算机应用与软件,2016,33(8):95 - 98.
中文标题
基于吉文斯变换提取ECG中J波信号
发表栏目
应用技术与研究
摘要点击数
486
英文标题
EXTRACTING J-WAVE FROM ECG BASED ON GIVENS TRANSFORMATION
作 者
朱贝贝 赵菊敏 李灯熬 Zhu Beibei Zhao Jumin Li Dengao
作者单位
太原理工大学信息工程学院 山西 太原 030024     
英文单位
College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,Shanxi,China     
关键词
J波信号 盲源分离 循环平稳性 吉文斯变换
Keywords
J-wave signal Blind source separation Cyclostationary Givens transformation
基金项目
国家自然科学基金面上项目(61371062)
作者资料
朱贝贝,硕士生,主研领域:生物医电信号处理。赵菊敏,教授。李灯熬,教授。 。
文章摘要
J波是心电图ECG(Electrocardiogram)上心室复极的一种新指标,它的出现可能预示着心脏发生室速甚至猝死的风险大大增加。近年来关于区分J波高危与良性形态的研究在医学上备受关注,但仅限于在临床上观察心电图上J波的时域波形,凭医生的经验作出判断,缺乏系统的评判标准。为了进一步研究J波信号的时频域特性,从ECG中准确地提取到J波信号是研究重点。分析了正常ECG信号及病变J波信号的循环平稳特性,采用以高阶循环平稳度HODCS(High-order Degree of Cyclostationarity)作为分离准则的盲源分离BSS(Blind Source Separation)算法,在利用小波包变换(Wavelet Packet Transform)定位S、T点和白化预处理之后使用吉文斯矩阵分离源信号。仿真结果表明,算法较经典Fast-ICA算法在提取J波信号的实际应用中有更好的分离性能。
Abstract
J-wave is used as a new diagnose index of Electrocardiogram (ECG) ventricular repolarisation.Its emergence is a presage of possible sharp increase in risks of ventricular tachycardia,ventricular fibrillation and sudden cardiac death.The studies on distinguishing high-risk and benign patterns of J-wave have attracted much concern in recent years,but they are only limited to the clinical observations of time domain waveform of J-wave on ECG,and the judgements made rely on doctors experiences,so there lacks the systematic evaluation criteria.For further studying the time-frequency domain characteristics of J-wave signal,our research focused on how to accurately extract J-wave from ECG.In this work we gave full consideration to the cyclostationary of normal ECG signal and diseased J-wave signal,and proposed a blind source separation (BSS) algorithm which uses high-order degree of cyclostationarity (HODCS) as the separation criterion.After positioning the ST-segment with wavelet packet transform (WPT) and then pre-processing with whitening,we used Givens transformation to separate source signals.Simulation results proved that this algorithm has better separation performance in practical application of J-wave extraction than classical Fast-ICA.
下载PDF全文