Non-invasive fetal ECG analysis

Abstract

Despite the important advances achieved in the field of adult electrocardiography signal processing, the analysis of the non-invasive fetal electrocardiogram (NI-FECG) remains a challenge. Currently no gold standard database exists which provides labelled FECG QRS complexes (and other morphological parameters), and publications rely either on proprietary databases or a very limited set of data recorded from few (or more often, just one) individuals. The PhysioNet/Computing in Cardiology Challenge 2013 enables to tackle some of these limitations by releasing a set of NI-FECG data publicly to the scientific community in order to evaluate signal processing techniques for NI-FECG extraction. The Challenge aim was to encourage development of accurate algorithms for locating QRS complexes and estimating the QT interval in noninvasive FECG signals. Using carefully reviewed reference QRS annotations and QT intervals as a gold standard, based on simultaneous direct FECG when possible, the Challenge was designed to measure and compare the performance of participants’ algorithms objectively. Multiple challenge events were designed to test basic FHR estimation accuracy, as well as accuracy in measurement of inter-beat (RR) and QT intervals needed as a basis for derivation of other FECG features. This editorial reviews the background issues, the design of the Challenge, the key achievements, and the follow-up research generated as a result of the Challenge, published in the concurrent special issue of Physiological Measurement.

An adaptive integrated algorithm for noninvasive fetal ECG separation and noise reduction based on ICA-EEMD-WS

“…Currently, significant effort is being placed on improving signal-to-noise ratio (SNR) of FECG signal, which includes removing the MECG complexes, reducing the effects of motion artifact, muscle noise and power line interface, and then enhancing the quality of the FECG before they can be consistently detected [7,15,32,36]. Recently, the PhysioNet/Computing in Cardiology Challenge 2013 (PCCC2013) has provided valuable data sets for the researchers.…”

“…Recently, the PhysioNet/Computing in Cardiology Challenge 2013 (PCCC2013) has provided valuable data sets for the researchers. The contest successfully attracted many experts in this field, and a large number of novel and effective methodologies were proposed [1,3,4,7,29,34]. In general, the most common approaches for separating and denoising the FECG from a mixed source signal can be divided into two groups: adaptive filtering and decomposition.…”

“…Both of the aforementioned algorithms obtained high-quality FECG. In general, adaptive filters are simple and effective algorithms, and they have played an important role in FECG extraction recently [4,7].…”

Extract Fetal ECG from Single-Lead Abdominal ECG by De-Shape Short Time Fourier Transform and Nonlocal Median

“…As is summarized in Clifford et al [5], most methods take the following four steps to study the aECG: first, pre-process the aECG; second, estimate the maECG; third, remove the maECG from the aECG; fourth, post-process the remainder to obtain the fECG and/or estimate the R peaks and hence the fIHR. In short, the maECG is removed first so that the fECG could be analyzed from the remainder.…”

Quantification of Feto-Maternal Heart Rate from Abdominal ECG Signal Using Empirical Mode Decomposition for Heart Rate Variability Analysis

“…Considering the above limitations, a non-invasive alternative for FHR monitoring is abdominal ECG (aECG) recordings. In the aECG method, electrodes are placed on the mother's abdomen, which acquire a composite ECG signal, which consists of both fetal ECG (fECG) and maternal ECG (mECG) [5] along with other artifacts like muscle contraction noise, power line interferences, various electrical noise, fetal movement and uterine contractions [6]. The interpretation of morphology, magnitude and frequency of fECG helps physicians to identify overall fetal health state [7]; however, the magnitude of fECG signals is usually overwhelmed by the magnitude of mECG signals, and significant frequency overlaps with other artifacts make the Signal to Noise Ratio (SNR) of the fECG signal very low.…”