de Araujo Draulio B, Barros Allan Kardec, Estombelo-Montesco Carlos, Zhao Hui, da Silva Filho A C Roque, Baffa Oswaldo, Wakai Ronald, Ohnishi Noboru
Department of Physics and Mathematics, FFCLRP, University of Sao Paulo, Ribeirao Preto, SP, Brazil.
Phys Med Biol. 2005 Oct 7;50(19):4457-64. doi: 10.1088/0031-9155/50/19/002. Epub 2005 Sep 13.
Fetal magnetocardiography (fMCG) has been extensively reported in the literature as a non-invasive, prenatal technique that can be used to monitor various functions of the fetal heart. However, fMCG signals often have low signal-to-noise ratio (SNR) and are contaminated by strong interference from the mother's magnetocardiogram signal. A promising, efficient tool for extracting signals, even under low SNR conditions, is blind source separation (BSS), or independent component analysis (ICA). Herein we propose an algorithm based on a variation of ICA, where the signal of interest is extracted using a time delay obtained from an autocorrelation analysis. We model the system using autoregression, and identify the signal component of interest from the poles of the autocorrelation function. We show that the method is effective in removing the maternal signal, and is computationally efficient. We also compare our results to more established ICA methods, such as FastICA.
胎儿心动图描记术（fMCG）作为一种可用于监测胎儿心脏各种功能的非侵入性产前技术，已在文献中被广泛报道。然而，fMCG信号的信噪比（SNR）往往较低，并且受到来自母亲心动图信号的强烈干扰。即使在低信噪比条件下，一种有前景的高效信号提取工具是盲源分离（BSS），即独立成分分析（ICA）。在此，我们提出一种基于ICA变体的算法，其中利用自相关分析获得的时间延迟来提取感兴趣的信号。我们使用自回归对系统进行建模，并从自相关函数的极点识别感兴趣的信号成分。我们表明该方法在去除母体信号方面是有效的，并且计算效率高。我们还将我们的结果与更成熟的ICA方法（如FastICA）进行了比较。
