基于心电信号提取呼吸信号的算法研究

Algorithm for extracting respiratory signals based on electrocardiogram signals

目的:为了实现医用心电监护仪器对多种参数的检测,减少设备的复杂性和降低患者的不适感,基于呼吸运动对心电信号影响的理论依据,提出一种从心电信号提取呼吸信息新算法。方法:运用Pan&Tompkins检测心电信号的R波和S波特征点,先后利用三次样条插值法和重采样法分别对此两路特征点进行处理,得到在相同位置采样拟合的R波序列和S波序列,选用小波变换理论重构一路呼吸信号序列,最后将处理得到的R波序列、S波序列、重构的呼吸信号序列和原信号4路信号序列构成混合矩阵,经独立分量分析(ICA)方法分离得到两路包含呼吸信息的源信号序列(Z1序列和Z2序列)。运用MATLAB软件对该算法的处理结果进行验证,并与相关的研究方法相比较。结果:在时域上对比统计人体每分钟呼吸次数,误差较小。经ICA方法提取出的两路源信号序列与其它呼吸信号波形有着良好的相关性,其平均相似度达到95.94%以上。结论:本研究提出的心电信号算法能够满足呼吸参数检测的需求,该算法是有效的。

Objective To propose a new algorithm of electrocardiogram(ECG)-derived respiratory signals detection for achieving the detection of a variety of physiological parameters, reducing the complexity of equipment as well as relieving the discomfort of patients. Methods The Pan & Tompkins algorithm was used to detect the feature points of R and S waves in ECG signals.Subsequently, the obtained feature points were processed with cubic spline interpolation and re-sampling methods for obtaining the fitted R and S wave sequences at the same positions. The wavelet transform method was applied to reconstruct a respiratory signal sequence. Finally, a hybrid matrix composed of R wave sequence, S wave sequence, the reconstructed respiratory signal sequence and the original signal sequence was formed and then processed with independent component analysis to extract two source signal sequences with respiratory information, namely Z1 sequence and Z2 sequence. MATLAB software was used to verify the results of the proposed algorithm and compare the results with those obtained with other algorithms. Results The error was smaller when detecting the number of breaths per minute of a human body in time domain. The two source signal sequences extracted with independent component analysis had good correlations with other respiratory waveforms, with an average similarity higher than 95.94%. Conclusion The proposed algorithm can meet the needs of respiratory parameter detection and is proved to be effective.