摘要
T波交替是预测恶性心律失常和心脏猝死危险性的一个无创的临床指标。针对目前T波交替(TWA)量化分析方法的不足,提出基于时频分析和相关分析相结合的T波交替联合分析算法:通过对预处理过的心电信号的连续小波变换,计算T波序列在指定时频窗内的能量谱,并对其分段统一编秩,再运用秩和检验方法定性判断T波交替;在此基础上,对T波交替的信号段进行T波的时域相关分析,确定交替的频次和程度,提取时域量化指标,为猝死和失常分析提供更丰富的时频特征信息。仿真实验表明,该算法对TWA检测的平均灵敏性达91.2%,在30 dB及以上信噪比的情况下,TWA被100%正确检测。临床数据检测结果表明,本算法对TWA幅值检测结果和临床应用的谱方法检测结果的相关系数达0.96。
T-wave alternant（TWA） is regarded as an independents and noninvasive indicator of sudden cardiac death（SCD）.A novel algorithm of quantitative estimation of TWA based on the joint analysis of time-frequency method and correlation technique was presented in this paper.The preprocessed electrocardiogram（ECG） was first applied with continuous wavelet transform,and then the T wave sequences were extracted and the total energy of each T wave contained within the selected time-frequency region was calculated.The rank-sum test was applied to the ranked energy sequences of T waves to detect TWA qualitatively.Then the ECG with TWA was analyzed with correlation method in temporal domain.And the alternant frequency and level were calculated,which provides more information in SCD and ventricular arrhythmia analysis.Simulation result proved a mean sensitivity of 91.2% in detecting TWA,and if the SNR was not less than 30 dB,the accuracy rate of detection by the algorithm achieved 100%.The clinical experiment data showed that the results from this method vs.spectral method have the correlation coefficients of 0.96.
出处
《中国生物医学工程学报》
CAS
CSCD
北大核心
2011年第4期487-493,共7页
Chinese Journal of Biomedical Engineering
基金
国家自然科学基金(60901027)
广东省教育厅广东高校优秀青年创新人才培育资助项目(LYM08050)
关键词
T波交替
连续小波变换
秩和检验
相关分析
量化评估
T wave alternans
continuous wavelet transform
rank-sum test
correlation analysis
quantitive estimation
