Echocardiographic assessment of fetal arrhythmias

Diagnosis of fetal arrhythmia in echocardiography imaging using deep learning with cyclic loss

Background

Fetal arrhythmias frequently co-occur with congenital heart disease in fetuses. The peaks observed in M-mode fetal echocardiograms serve as pivotal diagnostic markers for fetal arrhythmias. However, speckles, artifacts, and noise pose notable challenges for accurate image analysis. While current deep learning networks mainly overlook cardiac cyclic information, this study concentrated on the integration of such features, leveraging contextual constraints derived from cardiac cyclical features to improve diagnostic accuracy.

Methods

This study proposed a novel deep learning architecture for diagnosing fetal arrhythmias. The architecture presented a loss function tailored to the cardiac cyclical information and formulated a diagnostic algorithm for classifying fetal arrhythmias. The training and validation processes utilized a dataset comprising 4440 patches gathered from 890 participants.

Results

Incorporating cyclic loss significantly enhanced the performance of deep learning networks in predicting peak points for diagnosing fetal arrhythmia, resulting in improvements ranging from 7.11% to 14.81% in F1-score across different network combinations. Particularly noteworthy was the 18.2% improvement in the F1-score for the low-quality group. Additionally, the precision of diagnosing fetal arrhythmia across four categories exhibited improvement, with an average improvement rate of 20.6%.

Conclusion

This study introduced a cyclic loss mechanism based on the cardiac cycle information. Comparative evaluations were conducted using baseline methods and state-of-the-art deep learning architectures with the fetal echocardiogram dataset. These evaluations demonstrated the proposed framework's superior accuracy in diagnosing fetal arrhythmias. It is also crucial to note that further external testing is essential to assess the model's generalizability and clinical value.

Measurement of fetal atrioventricular intervals in pregnant women with anti-SSA/Ro antibodies

PURPOSE

The objective of our study was to compare and consider reference values of fetal atrioventricular (AV) intervals as measured by four different pulsed Doppler wave techniques (left ventricular inflow/outflow [LV in/out], pulmonary vein/pulmonary artery [PV/PA], innominate vein/ascending aorta [InnV/AA], and supra vena cava/ascending aorta [SVC/AA]) in pregnant women with anti-SSA/Ro antibodies.

METHODS

Between March 2014 and September 2020, 52 pregnant women with anti-SSA antibodies were enrolled. No bradyarrhythmia was observed in the group. A pulsed Doppler examination of the fetal heart was performed to obtain measurements of the mechanical Doppler AV interval. Doppler measurements were performed using four methods: LV in/out, PV/PA, InnV/AA, and SVC/AA. A statistical analysis was performed to examine the mean, standard deviation, significant difference, and correlation of the four methods. The detection rate of each method was also calculated.

RESULTS

There was no significant difference in the AV intervals between any of the four methods. There was also a positive correlation in the AV intervals of each of the four methods. The fetal heart rate and AV interval showed no correlation. The gestational age and AV interval also showed no correlation. The detection rate was highest for LV in/out (62.6%, 95% confidence interval: 56.5-68.4).

CONCLUSION

All four pulsed Doppler methods are useful for measuring AV intervals. The most practical method is LV in/out.

Detection of fetal arrhythmias in non-invasive fetal ECG recordings using data-driven entropy profiling

Objective:Fetal arrhythmias are a life-threatening disorder occurring in up to 2% of pregnancies. If identified, many fetal arrhythmias can be effectively treated using anti-arrhythmic therapies. In this paper, we present a novel method of detecting fetal arrhythmias in short length non-invasive fetal electrocardiography (NI-FECG) recordings.Approach:Our method consists of extracting a fetal heart rate (FHR) time series from each NI-FECG recording and computing an entropy profile using a data-driven range of the entropy tolerance parameter r. To validate our approach, we apply our entropy profiling method to a large clinical data set of 318 NI-FECG recordings.Main Results:We demonstrate that our method (TotalSampEn) provides strong performance for classifying arrhythmic fetuses (AUC of 0.83) and outperforms entropy measures such as SampEn (AUC of 0.68) and FuzzyEn (AUC of 0.72). We also find that NI-FECG recordings incorrectly classified using the investigated entropy measures have significantly lower signal quality, and that excluding recordings of low signal quality (13.5% of recordings) increases the classification accuracy of TotalSampEn (AUC of 0.90).Significance:The superior performance of our approach enables automated detection of fetal arrhythmias and warrants further investigation in a prospective clinical trial.

A non-invasive multimodal foetal ECG-Doppler dataset for antenatal cardiology research

Non-invasive foetal electrocardiography (fECG) continues to be an open topic for research. The development of standard algorithms for the extraction of the fECG from the maternal electrophysiological interference is limited by the lack of publicly available reference datasets that could be used to benchmark different algorithms while providing a ground truth for foetal heart activity when an invasive scalp lead is unavailable. In this work, we present the Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research (NInFEA), the first open-access multimodal early-pregnancy dataset in the field that features simultaneous non-invasive electrophysiological recordings and foetal pulsed-wave Doppler (PWD). The dataset is mainly conceived for researchers working on fECG signal processing algorithms. The dataset includes 60 entries from 39 pregnant women, between the 21st and 27th week of gestation. Each dataset entry comprises 27 electrophysiological channels (2048 Hz, 22 bits), a maternal respiration signal, synchronised foetal trans-abdominal PWD and clinical annotations provided by expert clinicians during signal acquisition. MATLAB snippets for data processing are also provided.

Recovery of the fetal electrocardiogram for morphological analysis from two trans-abdominal channels via optimal shrinkage

OBJECTIVE

We propose a novel algorithm to recover fetal electrocardiogram (ECG) for both the fetal heart rate analysis and morphological analysis of its waveform from two or three trans-abdominal maternal ECG channels.

APPROACH

We design an algorithm based on the optimal-shrinkage under the wave-shape manifold model. For the fetal heart rate analysis, the algorithm is evaluated on publicly available database, 2013 PhyioNet/Computing in Cardiology Challenge, set A (CinC2013). For the morphological analysis, we analyze CinC2013 and another publicly available database, non-invasive fetal ECG arrhythmia database (nifeadb), and propose to simulate semi-real databases by mixing the MIT-BIH normal sinus rhythm database and MITDB arrhythmia database.

MAIN RESULTS

For the fetal R peak detection, the proposed algorithm outperforms all algorithms under comparison. For the morphological analysis, the algorithm provides an encouraging result in recovery of the fetal ECG waveform, including PR, QT and ST intervals, even when the fetus has arrhythmia, both in real and simulated databases.

SIGNIFICANCE

To the best of our knowledge, this is the first work focusing on recovering the fetal ECG for morphological analysis from two or three channels with an algorithm potentially applicable for continuous fetal electrocardiographic monitoring, which creates the potential for long term monitoring purpose.

Measurement of fetal atrioventricular time intervals: A comparison of 3 spectral Doppler techniques

OBJECTIVE

To establish nomograms for fetal atrioventricular (AV) time intervals assessed by 3 different pulsed-wave Doppler techniques: left ventricular inflow and outflow tracts (LV in/out), superior vena cava and ascending aorta (SVC/AA), and pulmonary artery and pulmonary vein (PA/PV).

METHODS

A cross-sectional study was performed in 311 normal fetuses divided into 5 groups between 16 and 38 weeks. Pulsed-wave Doppler-derived AV intervals were measured by interrogation of flow in LV in/out, SVC/AA, and PA/PV. Linear regression analyses were performed to examine correlations with gestational age (GA) and fetal heart rate (FHR). Intraclass correlation coefficients for reproducibility of each method were compared.

RESULTS

Pulmonary artery and pulmonary vein revealed the longest mean AV time intervals (P < .001). The AV intervals in all methods were positively correlated with GA (R²  = 0.20-0.36; P < .001) and negatively correlated with FHR (R²  = 0.09-0.19; P < .001). The SCV/AA time intervals demonstrated the weakest influence of FHR. For LV in/out, SVC/AA, and PA/PV, intraobserver and interobserver reliability coefficients showed excellent agreements (all intraclass correlation coefficients ≥ 0.80).

CONCLUSION

All pulsed-wave Doppler-derived AV time intervals increased with advancing GA and decreased with increasing FHR. Fetal AV interval measurements can be obtained in a clinically viable fashion with excellent reproducibility.

Effectiveness of sotalol as first-line therapy for fetal supraventricular tachyarrhythmias

Fetal supraventricular tachycardia (SVT) and atrial flutter (AF) can be associated with significant morbidity and mortality. Digoxin is often used as first-line therapy but can be ineffective and is poorly transferred to the fetus in the presence of fetal hydrops. As an alternative to digoxin monotherapy, we have been using sotalol at presentation in fetuses with SVT or AF with, or at risk of, developing hydrops to attempt to achieve more rapid control of the arrhythmia. The present study was a retrospective review of the clinical, echocardiographic, and electrocardiographic data from all pregnancies with fetal tachycardia diagnosed and managed at a single center from 2004 to 2008. Of 29 affected pregnancies, 21 (16 SVT and 5 AF) were treated with sotalol at presentation, with or without concurrent administration of digoxin. Of the 21, 11 (6 SVT and 5 AF) had resolution of the tachycardia within 5 days (median 1). Six others showed some response (less frequent tachycardia, rate slowing, resolution of hydrops) without complete conversion. In 1 fetus with a slow response, the mother chose pregnancy termination. The 5 survivors with a slow response were all difficult to treat postnatally, including 1 requiring radiofrequency ablation as a neonate. One fetus developed blocked atrial extrasystoles after 1 dose of sotalol and was prematurely delivered for fetal bradycardia. Three grossly hydropic fetuses with SVT showed no response and died within 1 to 3 days of treatment. In conclusion, transplacental sotalol, alone or combined with digoxin, is effective for the treatment of fetal SVT and AF, with an 85% complete or partial response rate in our series.

Ultrasonographic diagnosis of delayed atrioventricular conduction during fetal life: a reliability study

OBJECTIVE

The objective of the study was to evaluate the reliability of the 2 most commonly used ultrasonographic approaches for monitoring fetal atrioventricular conduction time (AVCT): (1) superior vena cava/ascending aorta (SVC/AA), and (2) left ventricular inflow/outflow tract (LVI/O) Doppler recordings.

STUDY DESIGN

Echographic studies from fetuses followed up for first-degree atrioventricular block (AVB-1) between 1998 and 2008 were reviewed. The ability to identify atrial contractions in the same fetuses by the SVC/AA and LVI/O approaches was analyzed.

RESULTS

Sixty-six studies of 13 fetuses with AVB-1 were available. Atrial contractions were visible in all SVC/AA studies. With the LVI/O approach, atrial contractions could not be identified in 26 studies (39%). AVCT delay was significantly greater in the nonidentifiable compared with the identifiable atrial contraction group (P < .001). Differences in heart rate and gestational age were not significant.

CONCLUSION

The LVI/O is unsuitable for prenatal screening of conduction system anomalies.

Fetal Arrhythmias: A Clinical Review

Fetal rhythm abnormalities occur in 2% of pregnancies. They are usually identified by the obstetrician or midwifes after 20 weeks. There are four different methods used to assess fetal arrhythmias: scalp electrodes attached to electrocardiographic recordings, magnetocardiography (FMCG), fetal electrocardiographic recordings from the maternal abdomen, and fetal echocardiography (M-mode, pulsed-Doppler, Tissue-Doppler). In everyday practice the Doppler method was found to be the most useful method in the diagnosis and therapy of fetal arrhythmias. Doppler derived mechanical PR interval raised the possibility of refining the prenatal diagnosis of AV conduction abnormalities. A PR interval of >150 ms on Doppler, FMCG or postnatal ECG has been determined to be prolonged. Extrasystoles are most common cause of fetal arrhythmias, and are most often premature atrial contractions (PACs), what are usually identified in third trimester fetuses and their frequency may be highly variable. These are usually benign, resolving just before or shortly after birth. The follow-up is necessary, because some (1-3%) of affected fetuses have intermittent runs of supraventricular tachycardia. Ventricular tachycardia is rare during fetal life. With echocardiography in the setting of fetal tachycardia the findings of atrioventricular dissociation with a ventricular rate that is faster than the atrial rate suggests ventricular tachycardia. If there is 1:1 retrograde conduction it is impossible to distinguish between ventricular and supraventricular tachycardia. Atrial flutter accounted for 26.2% of all cases of fetal tachyarrhythmias and supraventricular tachycardia for 73.2%. Fouron and coworkers proposed to plan the management of the fetal tachyarrhythmia based on analysis of pulsed-Doppler recordings of fetal heart's blood flow. They determined short V-A tachycardia, when V-A (ventriculoatrial period) was shorter than AV (atrio-ventricular period) period. In the therapy of fetal supraventricular tachycardia there are different protocols, the most commonly used drugs are: digoxin, sotalol, amiodarone, flecainide. Persistent fetal sinus bradycardia is a rare condition and has been reported with central nervous system abnormalities, maternal treatment with beta blockers, excessive vagal tone, hydrops, long QT syndrome, intrauterine growth retardation and could be a sign of maternal anti-SSA/Ro antibodies. Prenatal sinus bradycardia or recognition of 2nd degree AV block may lead to early detection and treatment of long QT syndrome. Early detection of incomplete AV block, in cases of maternal anti SSA, SSB autoantibodies, successfully identifies a group at highest risk developing permanent AV block. The anti-inflammatory effects of dexamethasone might have interrupted on-going damage of the conduction system secondary to maternal autoantibodies. If the fetal arrhythmia resulted fetal hydrops, the mortality is high and the risk of late neurological morbidity must be taken into consideration. As a result of close follow-up, transplacentar treatment and well-organized perinatal management, the survival of sustained fetal arrhythmia significantly improved (50% versus 15%).

Overview of fetal arrhythmias

PURPOSE OF REVIEW

Though fetal arrhythmias account for a small proportion of referrals to a fetal cardiologist, they may be associated with significant morbidity and mortality. The present review outlines the current literature with regard to the diagnosis and, in brief, some management strategies in fetal arrhythmias.

RECENT FINDINGS

Advances in echocardiography have resulted in significant improvements in our ability to elucidate the mechanism of arrhythmia at the bedside. At the same time, magnetocardiography is broadening our understanding of mechanisms of arrhythmia especially as it pertains to ventricular arrhythmias and congenital heart block. It provides a unique window to study electrical properties of the fetal heart, unlike what has been available to date. Recent reports of bedside use of fetal ECG make it a promising new technology. Fetal magnetocardiography is also developing. The underlying mechanisms resulting in immune-mediated complete heart block in a small subset of 'at-risk' fetuses is under investigation.

SUMMARY

There have been great strides in noninvasive diagnosis of fetal arrhythmias. However, we still need to improve our knowledge of the electromechanical properties of the fetal heart as well as the mechanisms of arrhythmia to further improve outcomes. Multiinstitutional collaborative studies are needed to help answer some of the questions regarding patient, drug selection and management algorithms.