Giovanna Alleva

Fetal magnetocardiographic mapping using independent component analysis

Fetal magnetocardiography (fMCG) is the only noninvasive technique allowing effective assessment of fetal cardiac electrical activity during the prenatal period. The reconstruction of reliable magnetic field mapping associated with fetal heart activity would allow three-dimensional source localization. The efficiency of independent component analysis (ICA) in restoring reliable fetal traces from multichannel fMCG has already been demonstrated. In this paper, we describe a method of reconstructing a complete set of fetal signals hidden in multichannel fMCG preserving their correct spatial distribution, waveform, polarity and amplitude. Fetal independent components, retrieved with an ICA algorithm (FastICA), were interpolated (fICI method) using information gathered during FastICA iterations. The restored fetal signals were used to reconstruct accurate magnetic mapping for every millisecond during the average beat. The procedure was validated on fMCG recorded from the 22nd gestational week onward with a multichannel MCG system working in a shielded room. The interpolated traces were compared with those obtained with a standard technique, and the consistency of fetal mapping was checked evaluating source localizations relative to fetal echocardiographic information. Good magnetic field distributions during the P-QRS-T waves were attained with fICI for all gestational periods; their reliability was confirmed by three-dimensional source localizations.

Performance comparison of independent component analysis algorithms for fetal cardiac signal reconstruction: a study on synthetic fMCG data.

Independent component analysis (ICA) algorithms have been successfully used for signal extraction tasks in the field of biomedical signal processing. We studied the performances of six algorithms (FastICA, CubICA, JADE, Infomax, TDSEP and MRMI-SIG) for fetal magnetocardiography (fMCG). Synthetic datasets were used to check the quality of the separated components against the original traces. Real fMCG recordings were simulated with linear combinations of typical fMCG source signals: maternal and fetal cardiac activity, ambient noise, maternal respiration, sensor spikes and thermal noise. Clusters of different dimensions (19, 36 and 55 sensors) were prepared to represent different MCG systems. Two types of signal-to-interference ratios (SIR) were measured. The first involves averaging over all estimated components and the second is based solely on the fetal trace. The computation time to reach a minimum of 20 dB SIR was measured for all six algorithms. No significant dependency on gestational age or cluster dimension was observed. Infomax performed poorly when a sub-Gaussian source was included; TDSEP and MRMI-SIG were sensitive to additive noise, whereas FastICA, CubICA and JADE showed the best performances. Of all six methods considered, FastICA had the best overall performance in terms of both separation quality and computation times.

Performance comparison of six independent components analysis algorithms for fetal signal extraction from real fMCG data

In this study we compare the performance of six independent components analysis (ICA) algorithms on 16 real fetal magnetocardiographic (fMCG) datasets for the application of extracting the fetal cardiac signal. We also compare the extraction results for real data with the results previously obtained for synthetic data. The six ICA algorithms are FastICA, CubICA, JADE, Infomax, MRMI-SIG and TDSEP. The results obtained using real fMCG data indicate that the FastICA method consistently outperforms the others in regard to separation quality and that the performance of an ICA method that uses temporal information suffers in the presence of noise. These two results confirm the previous results obtained using synthetic fMCG data. There were also two notable differences between the studies based on real and synthetic data. The differences are that all six ICA algorithms are independent of gestational age and sensor dimensionality for synthetic data, but depend on gestational age and sensor dimensionality for real data. It is possible to explain these differences by assuming that the number of point sources needed to completely explain the data is larger than the dimensionality used in the ICA extraction.

Simultaneous monitoring of separate fetal magnetocardiographic signals in twin pregnancy.

Fetal magnetocardiography (fMCG) allows the non-invasive recording of fetal cardiac electrical activity with increasing efficacy as gestation progresses. Many reports on the successful extraction of reliable fetal magnetocardiographic traces in singleton pregnancies exist in the literature, whereas there is only one report on the reconstruction of averaged fetal cardiac signals obtained in a twin pregnancy with the use of a double sensor array system. In this paper, we aimed at assessing the effectiveness of an ICA-based procedure to reconstruct the time course of fetal cardiac signals recorded with a single-shot multi-channel fMCG device in an uncomplicated twin pregnancy at 27 weeks. The evaluation of heart rate and beats synchronicity permitted the differentiation of fetal components; the quality of reconstructed fetal signals allowed visual inspection on single cycles and the simultaneous monitoring of separate fetal heart rate patterns. The proposed technique might be applied in twin pregnancies not only to characterize fetal arrhythmias, but also in all cases of discordant fetal growth, either in the case of intra-uterine growth retardation affecting one fetus, or in the case of twin-twin transfusion syndrome, a life-threatening condition where both fetuses are at risk of heart failure.

Beat-to-beat estimate of fetal cardiac time intervals using magnetocardiography: longitudinal charts of normality ranges and individual trends

Background.  Fetal magnetocardiography (fMCG) records fetal cardiac electro‐physiological activity during the second half of gestation. We aimed at assessing normality values, related variability, and trends of fetal cardiac time intervals (fCTI) evaluated longitudinally from beat‐to‐beat fMCG analysis in uncomplicated pregnancies.

Entropy-based automated classification of independent components separated from fMCG

Fetal magnetocardiography (fMCG) is a noninvasive technique suitable for the prenatal diagnosis of the fetal heart function. Reliable fetal cardiac signals can be reconstructed from multi-channel fMCG recordings by means of independent component analysis (ICA). However, the identification of the separated components is usually accomplished by visual inspection. This paper discusses a novel automated system based on entropy estimators, namely approximate entropy (ApEn) and sample entropy (SampEn), for the classification of independent components (ICs). The system was validated on 40 fMCG datasets of normal fetuses with the gestational age ranging from 22 to 37 weeks. Both ApEn and SampEn were able to measure the stability and predictability of the physiological signals separated with ICA, and the entropy values of the three categories were significantly different at p <0.01. The system performances were compared with those of a method based on the analysis of the time and frequency content of the components. The outcomes of this study showed a superior performance of the entropy-based system, in particular for early gestation, with an overall ICs detection rate of 98.75% and 97.92% for ApEn and SampEn respectively, as against a value of 94.50% obtained with the time-frequency-based system.

A method for the automatic reconstruction of fetal cardiac signals from magnetocardiographic recordings

Fetal magnetocardiography (fMCG) allows monitoring the fetal heart function through algorithms able to retrieve the fetal cardiac signal, but no standardized automatic model has become available so far. In this paper, we describe an automatic method that restores the fetal cardiac trace from fMCG recordings by means of a weighted summation of fetal components separated with independent component analysis (ICA) and identified through dedicated algorithms that analyse the frequency content and temporal structure of each source signal. Multichannel fMCG datasets of 66 healthy and 4 arrhythmic fetuses were used to validate the automatic method with respect to a classical procedure requiring the manual classification of fetal components by an expert investigator. ICA was run with input clusters of different dimensions to simulate various MCG systems. Detection rates, true negative and false positive component categorization, QRS amplitude, standard deviation and signal-to-noise ratio of reconstructed fetal signals, and real and per cent QRS differences between paired fetal traces retrieved automatically and manually were calculated to quantify the performances of the automatic method. Its robustness and reliability, particularly evident with the use of large input clusters, might increase the diagnostic role of fMCG during the prenatal period.

Fetal cardiac time intervals estimated on fetal magnetocardiograms: single cycle analysis versus average beat inspection.

Fetal cardiac time intervals (fCTI) are dependent on fetal growth and development, and may reveal useful information for fetuses affected by growth retardation, structural cardiac defects or long QT syndrome. Fetal cardiac signals with a signal-to-noise ratio (SNR) of at least 15 dB were retrieved from fetal magnetocardiography (fMCG) datasets with a system based on independent component analysis (ICA). An automatic method was used to detect the onset and offset of the cardiac waves on single cardiac cycles of each signal, and the fCTI were quantified for each heartbeat; long rhythm strips were used to calculate average fCTI and their variability for single fetal cardiac signals. The aim of this work was to compare the outcomes of this system with the estimates of fCTI obtained with a classical method based on the visual inspection of averaged beats. No fCTI variability can be measured from averaged beats. A total of 25 fMCG datasets (fetal age from 22 to 37 weeks) were evaluated, and 1768 cardiac cycles were used to compute fCTI. The real differences between the values obtained with a single cycle analysis and visual inspection of averaged beats were very small for all fCTI. They were comparable with signal resolution (+/-1 ms) for QRS complex and QT interval, and always <5 ms for the PR interval, ST segment and T wave. The coefficients of determination between the fCTI estimated with the two methods ranged between 0.743 and 0.917. Conversely, inter-observer differences were larger, and the related coefficients of determination ranged between 0.463 and 0.807, assessing the high performance of the automated single cycle analysis, which is also rapid and unaffected by observer-dependent bias.

Automatic detection of cardiac waves on fetal magnetocardiographic signals.

Fetal magnetocardiography (fMCG) provides fetal cardiac traces useful for the prenatal monitoring of fetal heart function. In this paper, we describe an analytical model (ACWD) for the automatic detection of cardiac waves boundaries that works on fetal signals reconstructed from fMCG by means of independent component analysis. ACWD was validated for 45 healthy and 4 arrhythmic fetuses ranging from 22 to 37 weeks; ACWD outcomes were compared with the estimates of three independent investigators. Descriptive statistics were used to assess correspondence between the outcomes of the automatic and manual approaches. The parametric two-tailed Pearson correlation test (alpha=0.01) was employed to quantify, by means of the coefficients of determination, the amount of common variation between the sequences of intervals quantified automatically and manually. ACWD performances on short and long rhythm strips were investigated. ACWD demonstrated to be a robust tool providing dependable estimates of cardiac intervals and their variability during the third gestational trimester also in case of fetal arrhythmias. SNR and stability of fetal traces were the factors limiting ACWD performances. ACWD computation time, which was approximately 1:600 with respect to the manual procedure, was comparable with the time required for fCTI estimation on averaged beats.