基于小波变换的心阻抗血流图特征点提取

Detection of Impedance Cardiorgrahy's Characteristic Points Based on Wavelet Transform

以小波变换的极大值在多尺度间的变化与信号的瞬变有直接关系为基础,由小波变换的奇异点(过零点、极值点等)来观测心阻抗微分波形,在感兴趣的尺度上找到模极值点和特征点的对应关系,从而完成对特征点的自动检测。采用临床医学ICG数据对算法进行了检测,实验结果表明:心阻抗微分信号中B点、X点的漏检率与采用传统的相干平均法相比有明显降低,说明小波变换技术是心阻抗微分信号分析与处理的又一可行有效的方法。

With observation that singularities of a multi-scale wavelet transform result are related to discontinuities of the signal, a novel wavelet transform based method was proposed for detection of characteristic points of the impedance cardiography. The position of the points was obtained with interest in the cardiography signal by analyzing the ICG in diffcrent scale. The algorithm was tested using and proved to have satisfactory results compared with coherent average method, which s is an effective technology of analyzing and processing impedance differential signals.