Non-invasive Fetal Electrocardiography Research Articles

Fetal ECG extraction from time-varying and low-rank noninvasive maternal abdominal recordings

Objective: Noninvasive fetal electrocardiography is emerging as a low-cost and high-accuracy technology for fetal cardiac monitoring. Signal processing techniques have been used over the past fifty years in this domain. The current major challenges of this domain, addressed in this study are (1) fetal electrocardiogram (fECG) extraction from few numbers of maternal abdominal channels in low signal-to-noise ratios; (2) online fECG extraction; (3) automatic and online signal quality assessment and channel selection; and (4) accurate and robust fetal R-peak detection and ECG parameter extraction. Approach: Based on the theory of cyclostationarity, auxiliary maternal ECG channel(s) are synthetically constructed and augmented with the input channels. The augmented data are used to develop a robust multichannel source separation algorithm for online/offline fECG extraction, from as few as a single channel, and an accurate fetal R-peak detector using a two-pass matched filter. Several robust signal quality indexes (SQI) and a voting strategy are also proposed for automatic fetal signal quality assessment. Main results: It is shown that the fECG and the fetal R-peaks can be accurately extracted from standard online available datasets, for which classical source separation methods (requiring many channels) had previously failed. The signal quality indexes fully automate the extraction and channel selection procedure. Finally, the proposed R-peak detector is highly robust to background noise and residual maternal R-peak components. Significance: The proposed methods for fECG extraction, R-peak detection and automatic channel selection are evaluated (visually and numerically), on two online available datasets and compared with recently developed algorithms. The proposed algorithm is statistically shown to outperform the benchmarks in terms of average and standard deviation.

The effects of asymmetric volume conductor modeling on non-invasive fetal ECG extraction

Objective: Non-invasive fetal electrocardiography (NI-FECG) shows promise for capturing novel physiological information that may indicate signs of fetal distress. However, significant deterioration in NI-FECG signal quality occurs during the presence of a highly non-conductive layer known as vernix caseosa which forms on the fetal body surface beginning in approximately the 28th week of gestation. This work investigates asymmetric modeling of vernix caseosa and other maternal–fetal tissues in accordance with clinical observations and assesses their impacts for NI-FECG signal processing. Approach: We develop a process for simulating dynamic maternal–fetal abdominal ECG mixtures using a synthetic cardiac source model embedded in a finite element volume conductor. Using this process, changes in NI-FECG signal morphology are assessed in an extensive set of finite element models including spatially variable distributions of vernix caseosa. Main results: Our simulations show that volume conductor asymmetry can result in over 70% error in the observed T/QRS ratio and significant changes to signal morphology compared to a homogeneous volume conductor model. Volume conductor effects must be considered when analyzing T/QRS ratios obtained via NI-FECG and should be considered in future algorithm benchmarks using simulated data. Significance: This work shows that without knowledge of the influence of volume conductor effects, clinical evaluation of the T/QRS ratio derived via NI-FECG should be avoided.

Acceptance of a new non-invasive fetal monitoring system and attitude for telemedicine approaches in obstetrics: a case-control study.

Reduction of maternal morbidity and mortality is a major worldwide objective anchored in the millennium goals of the United Nations. To improve fetal and maternal care, a constant attempt to discover groundbreaking technologies is ongoing. One approach is the enhancement of non-invasive fetal ECG devices. Most importantly, acceptance of new technologies by pregnant women is a prerequisite for successful implementation. This questionnaire-based study conducted at the University Hospital Heidelberg, Germany between May and June 2017 evaluates pregnant women's attitudes towards a new device for fetal ECG monitoring and its potential home usage. The study population was questioned after exposure to the Parides/Atlantis prototype (Nemo Healthcare, Veldhoven, The Netherlands), whereas the maternal and gestational age-matched control group was left to envision telemedical topics. The prototype and its potential usage in a clinical and telemedical setting was highly accepted, and its comfort and appearance satisfied participants. Its use caused significantly improved telemedical understanding as envision increased (p = 0.0015). Implementation and integration of telemedical devices into antenatal care was significantly preferred by the study group (p = 0.0011), though participants desire more specific features for their personal use. Optional home-based self-monitoring to reduce scheduled doctoral visits (p = 0.0004) as well as self-assessment prior to self-initiated, unscheduled consultation (p < 0.0001) could be affected positively by such a device. Furthermore, it could reduce face-to-face interaction with the care provider (p = 0.0163). The positive feedback on remote self-monitoring might open options for a more "patient as partners" oriented prenatal care in the future. Safety and reliability remain a major issue. More comprehensive studies with new technologies are needed to diligently ensure quality of care. Finally, results for new technologies must be communicated to pregnant women for their acceptance and usage of new devices.

Non-invasive fetal electrocardiography for the detection of fetal arrhythmias: Toward a fetal Holter

Non-invasive fetal ECG is a promising low-cost and non-invasive continuous fetal monitoring technique. In this work, we explore the potential of non-invasive fetal electrocardiography as a novel technique for fetal arrhythmias diagnoses. The population of this prospective study comprised 12 arrhythmic fetus and 14 normal ones, according to a routine fetal echocardiography. We diagnosed the fetal cardiac rhythm of the 26 fetuses using the non-invasive fetal ECG and compared it to the echocardiographic reference diagnosis. For this purpose, extraction of the non-invasive fetal ECG and detection of abnormal rhythm events were performed automatically. Two perinatal cardiologists, blinded to the echocardiographic diagnosis, analyzed these events. Diagnosis based on the non-invasive fetal ECG recognized the following rhythm disorders: 6 cases with extrasystoles, 2 tachyarrhythmia, 1 bradyarrhythmia, 1 irregular atrial rhythm and 1 case with blocked P-waves ( Fig. 1 ). Non-invasive fetal ECG and fetal echocardiography agreed on all cases (26/26) on whether the fetus had normal rhythm or an arrhythmia. However, in one arrhythmic case the ECG miss-diagnosed the fetus with sinusal pauses although it had been diagnosed with atrial tachycardia with variable conduction by echocardiography. The non-invasive fetal ECG allowed to diagnose for fetal arrhythmias and in most cases provided refinement on the arrhythmias when compared to fetal echocardiography. This is mainly due to the longer recording time and the possibility to automate the detection of abnormal rhythm events. However, in some cases the low resolution of the P-wave in the ECG traces ( Fig. 2 ) made it difficult to resolve the mechanism behind the arrhythmia and in one case the low P-wave resolution leaded to an inaccurate diagnosis. The low-resolution of the extracted P-wave represents the main challenge for improving the diagnostic performance of fetal ECG. This paper is a proof of concept and a first step toward the creation of a fetal Holter ECG device. Such a device will offer new opportunities for fetal diagnosis and monitoring because it is low-cost, non-invasive, requires a minimal set-up, and can enable the detection of both persistent and intermittent fetal cardiac rhythm disorders.

Influence of Wilson Ce nter Terminal on fetal electrocardiography acquisition.

The acquisition of a standard 10-leads electrocardiography (ECG) is performed using the Wilson Center Terminal (WCT) reference with a normalized electrode positioning. However, in the case of non-invasive fetal ECG (fECG) acquisition, there is no standardization on the positioning of the electrodes on the abdomen and many authors suggest an acquisition with or without a WCT. The use of the WCT for the acquisition of the fetal heart rate (FHR) is not justified. The objective of this paper is to quantify the influence of this reference compared to a direct measurement. For this purpose, we developed a device allowing the acquisition of the fECG and compared the two configurations on 6 volunteer pregnants. The noise levels and the fetal QRS morphology were compared, showing no superiority of the WCT acquisition compared to a direct differential measurement.

A clustering-based method for single-channel fetal heart rate monitoring.

Non-invasive fetal electrocardiography (ECG) is based on the acquisition of signals from abdominal surface electrodes. The composite abdominal signal consists of the maternal electrocardiogram along with the fetal electrocardiogram and other electrical interferences. These recordings allow for the acquisition of valuable and reliable information that helps ensure fetal well-being during pregnancy. This paper introduces a procedure for fetal heart rate extraction from a single-channel abdominal ECG signal. The procedure is composed of three main stages: a method based on wavelet for signal denoising, a new clustering-based methodology for detecting fetal QRS complexes, and a final stage to correct false positives and false negatives. The novelty of the procedure thus relies on using clustering techniques to classify singularities from the abdominal ECG into three types: maternal QRS complexes, fetal QRS complexes, and noise. The amplitude and time distance of all the local maxima followed by a local minimum were selected as features for the clustering classification. A wide set of real abdominal ECG recordings from two different databases, providing a large range of different characteristics, was used to illustrate the efficiency of the proposed method. The accuracy achieved shows that the proposed technique exhibits a competitve performance when compared to other recent works in the literature and a better performance over threshold-based techniques.

A deep learning approach for fetal QRS complex detection

Objective: Non-invasive foetal electrocardiography (NI-FECG) has the potential to provide more additional clinical information for detecting and diagnosing fetal diseases. We propose and demonstrate a deep learning approach for fetal QRS complex detection from raw NI-FECG signals by using a convolutional neural network (CNN) model. The main objective is to investigate whether reliable fetal QRS complex detection performance can still be obtained from features of single-channel NI-FECG signals, without canceling maternal ECG (MECG) signals. Approach: A deep learning method is proposed for recognizing fetal QRS complexes. Firstly, we collect data from set-a of the PhysioNet/computing in Cardiology Challenge database. The sample entropy method is used for signal quality assessment. Part of the bad quality signals is excluded in the further analysis. Secondly, in the proposed method, the features of raw NI-FECG signals are normalized before they are fed to a CNN classifier to perform fetal QRS complex detection. We use precision, recall, F-measure and accuracy as the evaluation metrics to assess the performance of fetal QRS complex detection. Main results: The proposed deep learning method can achieve relatively high precision (75.33%), recall (80.54%), and F-measure scores (77.85%) compared with three other well-known pattern classification methods, namely KNN, naive Bayes and SVM. Significance: the proposed deep learning method can attain reliable fetal QRS complex detection performance from the raw NI-FECG signals without canceling MECG signals. In addition, the influence of different activation functions and signal quality assessment on classification performance are evaluated, and results show that Relu outperforms the Sigmoid and Tanh on this particular task, and better classification performance is obtained with the signal quality assessment step in this study.

The influence of betamethasone on fetal heart rate variability, obtained by non-invasive fetal electrocardiogram recordings

BackgroundBetamethasone is widely used to enhance fetal lung maturation in case of threatened preterm labour. Fetal heart rate variability is one of the most important parameters to assess in fetal monitoring, since it is a reliable indicator for fetal distress. AimTo describe the effect of betamethasone on fetal heart rate variability, by applying spectral analysis on non-invasive fetal electrocardiogram recordings. Study designProspective cohort study. SubjectsPatients that require betamethasone, with a gestational age from 24 weeks onwards. Outcome measuresFetal heart rate variability parameters on day 1, 2, and 3 after betamethasone administration are compared to a reference measurement. ResultsFollowing 68 inclusions, 12 patients remained with complete series of measurements and sufficient data quality. During day 1, an increase in absolute fetal heart rate variability values was seen. During day 2, a decrease in these values was seen. All trends indicate to return to pre-medication values on day 3. Normalised high- and low-frequency power show little changes during the study period. ConclusionsThe changes in fetal heart rate variability following betamethasone administration show the same pattern when calculated by spectral analysis of the fetal electrocardiogram, as when calculated by cardiotocography. Since normalised spectral values show little changes, the influence of autonomic modulation seems minor.

ECG Electrode Configuration to Extract Real Time FECG Signals

Monitoring fetal health early in pregnancy helps to avert possible fetal damage and even fetal death. Among various fetal monitoring techniques the invasive fetal scalp electrode method is the most accurate method at labor. However this method may cause risk to the baby if used for antepartum monitoring. The Non-invasive fetal ECG (NIFECG) is a simple and the next best method to extract fetal heart rate (FHR) before full term. NIFECG can extract the fetal morphology from the FECG signal which can aid in diagnosing fetal health ahead of delivery. This paper aims to locate the most optimum points to place the ECG electrodes on the maternal abdomen so as to get the best FECG signals. After reviewing few previous techniques, we proposed three schemes using a standard 12 lead ECG recorder. It was observed that the scheme 2 successfully displayed the average FHR as compared to the ultrasound report of a single subject. A further researching and experimental procedures are required in this abdominal ECG (aECG) placement electrode positioning schemes before a standardized electrode configuration which can be introduced in clinical practice.

Wavelet optimization for applying continuous wavelet transform to maternal electrocardiogram component enhancing**Project supported by the National Natural Science Foundation of China (Grant No. 61271079)

In the procedure of non-invasive fetal electrocardiogram (ECG) extraction, high-quality maternal R wave peak detection demands enhancing the maternal ECG component firstly. Among all the enhancing algorithms, the one based on the continuous wavelet transform (CWT) is very important and its effectiveness depends on the optimization of the used wavelet. However, up to now, there is still no clear conclusion on the optimal wavelet (including type and scale) for CWT to enhance the maternal ECG component of an abdominal ECG signal. To solve this problem, in this paper, we select several common used types of wavelets to carry out our research on what the optimal wavelets are. We first establish big-enough training datasets with different sampling rates and make a maternal QRS template for each signal in the training datasets. Second, for each type of selected wavelets, we find its optimal scale corresponding to each QRS template in a training dataset based on the principle of maximal correlation. Then calculating the average of all optimized wavelet scales results in the mean optimal wavelet of this type for the dataset. We use two original abdominal ECG databases to train and test the optimized mean optimal wavelets. The test results show that, as a whole, the mean optimal wavelets obtained are superior to the wavelets used in other publications for applying CWT to maternal ECG component enhancing.

A Combined Independent Source Separation and Quality Index Optimization Method for Fetal ECG Extraction from Abdominal Maternal Leads.

The non-invasive fetal electrocardiogram (fECG) technique has recently received considerable interest in monitoring fetal health. The aim of our paper is to propose a novel fECG algorithm based on the combination of the criteria of independent source separation and of a quality index optimization (ICAQIO-based). The algorithm was compared with two methods applying the two different criteria independently—the ICA-based and the QIO-based methods—which were previously developed by our group. All three methods were tested on the recently implemented Fetal ECG Synthetic Database (FECGSYNDB). Moreover, the performance of the algorithm was tested on real data from the PhysioNet fetal ECG Challenge 2013 Database. The proposed combined method outperformed the other two algorithms on the FECGSYNDB (ICAQIO-based: 98.78%, QIO-based: 97.77%, ICA-based: 97.61%). Significant differences were obtained in particular in the conditions when uterine contractions and maternal and fetal ectopic beats occurred. On the real data, all three methods obtained very high performances, with the QIO-based method proving slightly better than the other two (ICAQIO-based: 99.38%, QIO-based: 99.76%, ICA-based: 99.37%). The findings from this study suggest that the proposed method could potentially be applied as a novel algorithm for accurate extraction of fECG, especially in critical recording conditions.

Noninvasive Fetal Electrocardiography Part I: Pan-Tompkins' Algorithm Adaptation to Fetal R-peak Identification.

Background:Indirect fetal electrocardiography is preferable to direct fetal electrocardiography because of being noninvasive and is applicable also during the end of pregnancy, besides labor. Still, the former is strongly affected by noise so that even R-peak detection (which is essential for fetal heart-rate evaluations and subsequent processing procedures) is challenging. Some fetal studies have applied the Pan-Tompkins’ algorithm that, however, was originally designed for adult applications. Thus, this work evaluated the Pan-Tompkins’ algorithm suitability for fetal applications, and proposed fetal adjustments and optimizations to improve it. Method: Both Pan-Tompkins’ algorithm and its improved version were applied to the “Abdominal and Direct Fetal Electrocardiogram Database” and to the “Noninvasive Fetal Electrocardiography Database” of Physionet. R-peak detection accuracy was quantified by computation of positive-predictive value, sensitivity and F1 score. Results:When applied to “Abdominal and Direct Fetal Electrocardiogram Database”, the accuracy of the improved fetal Pan-Tompkins’ algorithm was significantly higher than the standard (positive-predictive value: 0.94 vs. 0.79; sensitivity: 0.95 vs. 0.80; F1 score: 0.94 vs. 0.79; P<0.05 in all cases) on indirect fetal electrocardiograms, whereas both methods performed similarly on direct fetal electrocardiograms (positive-predictive value, sensitivity and F1 score all close to 1). Improved fetal Pan-Tompkins’ algorithm was found to be superior to the standard also when applied to “Noninvasive Fetal Electrocardiography Database” (positive-predictive value: 0.68 vs. 0.55, P<0.05; sensitivity: 0.56 vs. 0.46, P=0.23; F1 score: 0.60 vs. 0.47, P=0.11).Conclusion:In indirect fetal electrocardiographic applications, improved fetal Pan-Tompkins’ algorithm is to be preferred over the standard, since it provides higher R-peak detection accuracy for heart-rate evaluations and subsequent processing.

Non-invasive Fetal ECG Signal Quality Assessment for Multichannel Heart Rate Estimation.

The noninvasive fetal ECG (NI-FECG) from abdominal recordings offers novel prospects for prenatal monitoring. However, NI-FECG signals are corrupted by various nonstationary noise sources, making the processing of abdominal recordings a challenging task. In this paper, we present an online approach that dynamically assess the quality of NI-FECG to improve fetal heart rate (FHR) estimation. Using a naive Bayes classifier, state-of-the-art and novel signal quality indices (SQIs), and an existing adaptive Kalman filter, FHR estimation was improved. For the purpose of training and validating the proposed methods, a large annotated private clinical dataset was used. The suggested classification scheme demonstrated an accuracy of Krippendorff's alpha in determining the overall quality of NI-FECG signals. The proposed Kalman filter outperformed alternative methods for FHR estimation achieving accuracy. The proposed algorithm was able to reliably reflect changes of signal quality and can be used in improving FHR estimation. NI-ECG signal quality estimation and multichannel information fusion are largely unexplored topics. Based on previous works, multichannel FHR estimation is a field that could strongly benefit from such methods. The developed SQI algorithms as well as resulting classifier were made available under a GNU GPL open-source license and contributed to the FECGSYN toolbox.

Extraction of fetal ECG signal by an improved method using extended Kalman smoother framework from single channel abdominal ECG signal.

This paper proposes a five-stage based methodology to extract the fetal electrocardiogram (FECG) from the single channel abdominal ECG using differential evolution (DE) algorithm, extended Kalman smoother (EKS) and adaptive neuro fuzzy inference system (ANFIS) framework. The heart rate of the fetus can easily be detected after estimation of the fetal ECG signal. The abdominal ECG signal contains fetal ECG signal, maternal ECG component, and noise. To estimate the fetal ECG signal from the abdominal ECG signal, removal of the noise and the maternal ECG component presented in it is necessary. The pre-processing stage is used to remove the noise from the abdominal ECG signal. The EKS framework is used to estimate the maternal ECG signal from the abdominal ECG signal. The optimized parameters of the maternal ECG components are required to develop the state and measurement equation of the EKS framework. These optimized maternal ECG parameters are selected by the differential evolution algorithm. The relationship between the maternal ECG signal and the available maternal ECG component in the abdominal ECG signal is nonlinear. To estimate the actual maternal ECG component present in the abdominal ECG signal and also to recognize this nonlinear relationship the ANFIS is used. Inputs to the ANFIS framework are the output of EKS and the pre-processed abdominal ECG signal. The fetal ECG signal is computed by subtracting the output of ANFIS from the pre-processed abdominal ECG signal. Non-invasive fetal ECG database and set A of 2013 physionet/computing in cardiology challenge database (PCDB) are used for validation of the proposed methodology. The proposed methodology shows a sensitivity of 94.21%, accuracy of 90.66%, and positive predictive value of 96.05% from the non-invasive fetal ECG database. The proposed methodology also shows a sensitivity of 91.47%, accuracy of 84.89%, and positive predictive value of 92.18% from the set A of PCDB.

Monitoring Fetal Heart Rate during Labor: A Comparison of Three Methods

The purpose of the study was to compare the accuracy of a noninvasive fetal heart rate monitor with that of ultrasound, using a fetal scalp electrode as the gold standard, in laboring women of varying body habitus, throughout labor and delivery. Laboring women requiring fetal scalp electrode were monitored simultaneously with the investigational device (noninvasive fetal ECG), ultrasound, and fetal scalp electrode. An algorithm extracted the fetal heart rate from the noninvasive fetal ECG signal. Each noninvasive device recording was compared with fetal scalp electrode with regard to reliability by positive percent agreement and accuracy by root mean squared error. Seventy-one women were included in this analysis. Positive percent agreement was 83.4 ± 15.4% for noninvasive fetal ECG and 62.4 ± 26.7% for ultrasound. The root mean squared error compared with fetal scalp electrode-derived fetal heart rate was 4.8 ± 2.0 bpm for noninvasive fetal ECG and 14.3 ± 8.2 bpm for ultrasound. The superiority of noninvasive fetal ECG was maintained for stages 1 and 2 of labor and increases in body mass index. Compared with fetal scalp electrode-derived fetal heart rate, noninvasive fetal ECG is more accurate and reliable than ultrasound for intrapartum monitoring for stages 1 and 2 of labor and is less affected by increasing maternal body mass index. This confirms the results of other workers in this field.

Energy and Quality Evaluation for Compressive Sensing of Fetal Electrocardiogram Signals.

This manuscript addresses the problem of non-invasive fetal Electrocardiogram (ECG) signal acquisition with low power/low complexity sensors. A sensor architecture using the Compressive Sensing (CS) paradigm is compared to a standard compression scheme using wavelets in terms of energy consumption vs. reconstruction quality, and, more importantly, vs. performance of fetal heart beat detection in the reconstructed signals. We show in this paper that a CS scheme based on reconstruction with an over-complete dictionary has similar reconstruction quality to one based on wavelet compression. We also consider, as a more important figure of merit, the accuracy of fetal beat detection after reconstruction as a function of the sensor power consumption. Experimental results with an actual implementation in a commercial device show that CS allows significant reduction of energy consumption in the sensor node, and that the detection performance is comparable to that obtained from original signals for compression ratios up to about 75%.

Fetal motion estimation from noninvasive cardiac signal recordings

Fetal motility is a widely accepted indicator of the well-being of a fetus. In previous research, it has be shown that fetal motion (FM) is coherent with fetal heart rate accelerations and an indicator for active/rest cycles of the fetus. The most common approach for FM and fetal heart rate (FHR) assessment is by Doppler ultrasound (DUS). While DUS is the most common approach for studying the mechanical activities of the heart, noninvasive fetal electrocardiogram (ECG) and magnetocardiogram (MCG) recording and processing techniques have been considered as a possible competitor (or complement) for the DUS.In this study, a fully automatic and robust framework is proposed for the extraction, ranking and alignment of fetal QRS-complexes from noninvasive fetal ECG/MCG. Using notions from subspace tracking, two measures, namely the actogram and rotatogram, are defined for fetal motion tracking. The method is applied to four fetal ECG/MCG databases, including twin MCG recordings. By defining a novel measure of causality, it is shown that there is significant coherency and causal relationship between the actogram/rotatogram and FHR accelerations/decelerations. Using this measure, it is shown that in many cases, the actogram and rotatogram precede the FHR variations, which supports the idea of motion-induced FHR accelerations/decelerations for these cases and raises attention for the non-motion-induced FHR variations, which can be associated to the fetal central nervous system developments. The results of this study can lead to novel perspectives of the fetal sympathetic and parasympathetic brain systems and future requirements of fetal cardiac monitoring.

Evaluation of the fetal QT interval using non-invasive fetal ECG technology

Non-invasive fetal electrocardiography (NI-FECG) is a promising alternative continuous fetal monitoring method that has the potential to allow morphological analysis of the FECG. However, there are a number of challenges associated with the evaluation of morphological parameters from the NI-FECG, including low signal to noise ratio of the NI-FECG and methodological challenges for getting reference annotations and evaluating the accuracy of segmentation algorithms. This work aims to validate the measurement of the fetal QT interval in term laboring women using a NI-FECG electrocardiogram monitor. Fetal electrocardiogram data were recorded from 22 laboring women at term using the NI-FECG and an invasive fetal scalp electrode simultaneously. A total of 105 one-minute epochs were selected for analysis. Three pediatric electrophysiologists independently annotated individual waveforms and averaged waveforms from each epoch. The intervals measured on the averaged cycles taken from the NI-FECG and the fetal scalp electrode showed a close agreement; the root mean square error between all corresponding averaged NI-FECG and fetal scalp electrode beats was 13.6 ms, which is lower than the lowest adult root mean square error of 16.1 ms observed in related adult QT studies. These results provide evidence that NI-FECG technology enables accurate extraction of the fetal QT interval.

An open-source framework for stress-testing non-invasive foetal ECG extraction algorithms

Over the past decades, many studies have been published on the extraction of non-invasive foetal electrocardiogram (NI-FECG) from abdominal recordings. Most of these contributions claim to obtain excellent results in detecting foetal QRS (FQRS) complexes in terms of location. A small subset of authors have investigated the extraction of morphological features from the NI-FECG. However, due to the shortage of available public databases, the large variety of performance measures employed and the lack of open-source reference algorithms, most contributions cannot be meaningfully assessed.This article attempts to address these issues by presenting a standardised methodology for stress testing NI-FECG algorithms, including absolute data, as well as extraction and evaluation routines. To that end, a large database of realistic artificial signals was created, totaling 145.8 h of multichannel data and over one million FQRS complexes. An important characteristic of this dataset is the inclusion of several non-stationary events (e.g. foetal movements, uterine contractions and heart rate fluctuations) that are critical for evaluating extraction routines. To demonstrate our testing methodology, three classes of NI-FECG extraction algorithms were evaluated: blind source separation (BSS), template subtraction (TS) and adaptive methods (AM). Experiments were conducted to benchmark the performance of eight NI-FECG extraction algorithms on the artificial database focusing on: FQRS detection and morphological analysis (foetal QT and T/QRS ratio).The overall median FQRS detection accuracies (i.e. considering all non-stationary events) for the best performing methods in each group were 99.9% for BSS, 97.9% for AM and 96.0% for TS. Both FQRS detections and morphological parameters were shown to heavily depend on the extraction techniques and signal-to-noise ratio. Particularly, it is shown that their evaluation in the source domain, obtained after using a BSS technique, should be avoided. Data, extraction algorithms and evaluation routines were released as part of the fecgsyn toolbox on Physionet under an GNU GPL open-source license. This contribution provides a standard framework for benchmarking and regulatory testing of NI-FECG extraction algorithms.

Fusion of detected multi-channel maternal electrocardiogram (ECG) R-wave peak locations.

BackgroundAlmost all promising non-invasive foetal ECG extraction methods involve accurately determining maternal ECG R-wave peaks. However, it is not easy to robustly detect accurate R-wave peaks of the maternal ECG component in an acquired abdominal ECG since it often has a low signal-to-noise ratio (SNR), sometimes containing a large foetal ECG component or other noises and interferences. This paper discusses, under the condition of acquiring multi-channel abdominal ECG signals, how to improve the robustness of maternal ECG R-wave peak detection.MethodsOn the basis of summarising the current single channel ECG R-wave peak detection methods, the paper proposed a specific fusion algorithm of detected multi-channel maternal ECG R-wave peak locations. The proposed entire algorithm was then tested using two databases; one database, created by us, was composed of 343 groups of 8-channel data collected from 78 pregnant women, and the other one, called the challenge database, was from the Physionet/Computing in Cardiology Challenge 2013, including 175 groups of 4-channel data. When using these databases, each group of data was classified into two parts, called the training part and the validation test part respectively; the training part was the first 8.192 s of each group of data and the validation test part was the next 8.192 s.ResultsTo show the results, three evaluation parameters—sensitivity (Se), positive predictive value (PPV) and F1—are used. The validation test results for the database we collected are Se = 99.93 %, PPV = 99.98 %, and F1 = 99.95 %, while the results for the challenge database are Se = 99.91 %, PPV = 99.86 %, and F1 = 99.88 %.ConclusionThe results of the test show that the robustness of our proposed whole fusion algorithm was superior to that of other outstanding algorithms for maternal R-wave detection, and is much better than that of single channel maternal R-wave detection algorithms.