Comparative Effectiveness of ICA and PCA in Extraction of Fetal ECG From Abdominal Signals: Toward Non-invasive Fetal Monitoring

Abstract

Non-adaptive signal processing methods have been successfully applied to extract fetal electrocardiograms (fECGs) from maternal abdominal electrocardiograms (aECGs); and initial tests to evaluate the efficacy of these methods have been carried out by using synthetic data. Nevertheless, performance evaluation of such methods using real data is a much more challenging task and has neither been fully undertaken nor reported in the literature. Therefore, in this investigation, we aimed to compare the effectiveness of two popular non-adaptive methods (the ICA and PCA) to explore the non-invasive (NI) extraction (separation) of fECGs, also known as NI-fECGs from aECGs. The performance of these well-known methods was enhanced by an adaptive algorithm, compensating amplitude difference and time shift between the estimated components. We used real signals compiled in 12 recordings (real01–real12). Five of the recordings were from the publicly available database (PhysioNet-Abdominal and Direct Fetal Electrocardiogram Database), which included data recorded by multiple abdominal electrodes. Seven more recordings were acquired by measurements performed at the Institute of Medical Technology and Equipment, Zabrze, Poland. Therefore, in total we used 60 min of data (i.e., around 88,000 R waves) for our experiments. This dataset covers different gestational ages, fetal positions, fetal positions, maternal body mass indices (BMI), etc. Such a unique heterogeneous dataset of sufficient length combining continuous Fetal Scalp Electrode (FSE) acquired and abdominal ECG recordings allows for robust testing of the applied ICA and PCA methods. The performance of these signal separation methods was then comprehensively evaluated by comparing the fetal Heart Rate (fHR) values determined from the extracted fECGs with those calculated from the fECG signals recorded directly by means of a reference FSE. Additionally, we tested the possibility of non-invasive ST analysis (NI-STAN) by determining the T/QRS ratio. Our results demonstrated that even though these advanced signal processing methods are suitable for the non-invasive estimation and monitoring of the fHR information from maternal aECG signals, their utility for further morphological analysis of the extracted fECG signals remains questionable and warrants further work.

Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal

“…Fetal monitoring during pregnancy is very important for identifying many factors that may negatively affect the health of the fetus, may prevent intrauterine death (most commonly occurring in the home environment), or permanent damage to the fetus [ 1 , 2 , 3 , 4 , 5 ]. These factors may not only be harmful to the health of the fetus, but also to the health of the mother.…”

“…Some issues, such as premature birth, hypoxia, or intrauterine retardation, are dangerous both for the fetus and for the mother. Fetal monitoring includes methods such as fetal electrocardiography (fECG) [ 1 , 2 , 3 , 6 , 7 , 8 , 9 , 10 , 11 ], fetal phonocardiography (fPCG) [ 4 , 5 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 ], fetal echocardiography (fECHO) [ 22 , 23 , 24 , 25 ], fetal magnetocardiography (fMCG) [ 26 , 27 , 28 , 29 , 30 ], and cardiotocography (CTG) [ 9 , 31 , 32 , 33 ], which is based on Doppler ultrasound. Each fetal monitoring method has its advantages as well as disadvantages, and it should be emphasized that, regarding long-term measurements, a 20-min measurement is relatively short in order to obtain real information about the condition of the fetus.…”

“…In Table 1 , we summarize the main information about each of the techniques showed in Figure 2 , such as the technical solution or the limitation of its usage in terms of gestation age. Moreover, Table 1 offers comparison of the most significant benefits and drawbacks of the EFM methods [ 1 , 2 , 4 , 16 , 37 , 38 , 39 ]. The summary implies that NI-fECG is a very promising method.…”

“…In fECG signal processing, these methods require a clean mECG signal from the mother’s chest and an aECG signal, which contains both maternal and fetal components and noises. Hybrid neural network (HNN), artificial neural networks (ANN), and techniques of adaptive neuro-fuzzy inference system (ANFIS) fall into nonlinear adaptive techniques and the methods based on the theory of Kalman filtering (KF), least mean squares algorithm (LMS), recursive least squares algorithm (RLS), and methods based on adaptive linear neuron (ADALINE) [ 1 , 2 , 3 , 6 ] fall into linear adaptive methods.…”

“…These methods are very fast and accurate, but their disadvantage is that they are time invariant in nature compared with the adaptive methods. For interference elimination in aECG signal, the non-adaptive methods use either single channel signal sources, which are implemented by many methodologies, or multichannel signal sources, which are processed by blind source separation techniques (BSS) [ 1 , 2 , 3 , 6 ].…”