Sharma Hemant, Sharma K K, Bhagat Om Lata
Department of Electronics & Communication, Malaviya National Institute. Of Technology, Jaipur 302017, India.
Department of Electronics & Communication, Malaviya National Institute. Of Technology, Jaipur 302017, India.
Comput Biol Med. 2015 Apr;59:80-86. doi: 10.1016/j.compbiomed.2015.01.024. Epub 2015 Feb 9.
In this paper a new technique for the extraction of respiratory signal from the single-lead ECG using generalized homomorphic filtering is presented. It is proposed to perform band pass filtering on the cepstrum of the ECG signal to extract the respiratory signal. For this study, transforms used in generalized homomorphic filtering are the discrete Fourier transform (DFT) and the discrete cosine transform (DCT). The performance of the ECG-derived respiration (EDR) signal obtained using the proposed method is compared with the reference respiratory signal in terms of the correlation, magnitude squared coherence coefficients and breath rate accuracy. It is observed from the comparisons that the EDR technique based on generalized homomorphic filtering using DFT performs better than the homomorphic filtering using DCT. The proposed EDR technique is also compared with the two well-known EDR techniques: principal component analysis and R peak amplitude algorithm. It is seen from the results that the proposed EDR technique (RDFT) performs significantly better than the R peak amplitude algorithm, but significant improvements are not observed when compared with the PCA based EDR technique.
本文提出了一种利用广义同态滤波从单导联心电图中提取呼吸信号的新技术。该技术建议对心电图信号的倒谱进行带通滤波以提取呼吸信号。在本研究中，广义同态滤波中使用的变换是离散傅里叶变换（DFT）和离散余弦变换（DCT）。将使用所提方法获得的心电图衍生呼吸（EDR）信号的性能与参考呼吸信号在相关性、幅度平方相干系数和呼吸率准确性方面进行比较。从比较中可以观察到，基于使用DFT的广义同态滤波的EDR技术比使用DCT的同态滤波表现更好。所提的EDR技术还与两种著名的EDR技术：主成分分析和R波峰值算法进行了比较。从结果可以看出，所提的EDR技术（RDFT）比R波峰值算法表现明显更好，但与基于主成分分析的EDR技术相比未观察到显著改进。
