Fetal Cardiac Screening Research Articles

Image Segmentation of the Ventricular Septum in Fetal Cardiac Ultrasound Videos Based on Deep Learning Using Time-Series Information.

Image segmentation is the pixel-by-pixel detection of objects, which is the most challenging but informative in the fundamental tasks of machine learning including image classification and object detection. Pixel-by-pixel segmentation is required to apply machine learning to support fetal cardiac ultrasound screening; we have to detect cardiac substructures precisely which are small and change shapes dynamically with fetal heartbeats, such as the ventricular septum. This task is difficult for general segmentation methods such as DeepLab v3+, and U-net. Hence, here we proposed a novel segmentation method named Cropping-Segmentation-Calibration (CSC) that is specific to the ventricular septum in ultrasound videos in this study. CSC employs the time-series information of videos and specific section information to calibrate the output of U-net. The actual sections of the ventricular septum were annotated in 615 frames from 421 normal fetal cardiac ultrasound videos of 211 pregnant women who were screened. The dataset was assigned a ratio of 2:1, which corresponded to a ratio of the training to test data, and three-fold cross-validation was conducted. The segmentation results of DeepLab v3+, U-net, and CSC were evaluated using the values of the mean intersection over union (mIoU), which were 0.0224, 0.1519, and 0.5543, respectively. The results reveal the superior performance of CSC.

Impact of maternal obesity on fetal cardiac screening: which follow-up strategy is cost-effective?

To perform a cost-effectiveness analysis of different follow-up strategies for non-obese and obese women who had incomplete fetal cardiac screening for major congenital heart disease (CHD). Three decision-analytic models, one each for non-obese, obese and Class-III-obese women, were developed to compare five follow-up strategies for initial suboptimal fetal cardiac screening. The five strategies were: (1) no follow-up ultrasound (US) examination but direct referral to fetal echocardiography (FE); (2) one follow-up US, then FE if fetal cardiac views were still suboptimal; (3) up to two follow-up US, then FE if fetal cardiac views were still suboptimal; (4) one follow-up US and no FE; and (5) up to two follow-up US and no FE. The models were designed to identify fetuses with major CHD in a theoretical cohort of 4 000 000 births in the USA. Outcomes related to neonatal mortality and neurodevelopmental disability were evaluated. A cost-effectiveness willingness-to-pay threshold was set at US$100 000 per quality-adjusted life year (QALY). Base-case and sensitivity analysis and Monte-Carlo simulation were performed. In our base-case models for all body mass index (BMI) groups, no follow-up US, but direct referral to FE led to the best outcomes, detecting 7%, 25% and 82% more fetuses with CHD in non-obese, obese and Class-III-obese women, respectively, compared with the baseline strategy of one follow-up US and no FE. However, no follow-up US, but direct referral to FE was above the US$100 000/QALY threshold and therefore not cost-effective. The cost-effective strategy for all BMI groups was one follow-up US and no FE. Both up to two follow-up US with no FE and up to two follow-up US with FE were dominated (being more costly and less effective), while one follow-up US with FE was over the cost-effectiveness threshold. One follow-up US and no FE was the optimal strategy in 97%, 93% and 86% of trials in Monte-Carlo simulation for non-obese, obese and Class-III-obese models, respectively. For both non-obese and obese women with incomplete fetal cardiac screening, the optimal CHD follow-up screening strategy is no further US and immediate referral to FE; however, this strategy is not cost-effective. Considering costs, one follow-up US and no FE is the preferred strategy. For both obese and non-obese women, Monte-Carlo simulations showed clearly that one follow-up US and no FE was the optimal strategy. Both non-obese and obese women with initial incomplete cardiac screening examination should therefore be offered one follow-up US. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

VP17.02: A five‐year audit on fetal cardiac screening from a busy district general hospital

VP17.02: A five‐year audit on fetal cardiac screening from a busy district general hospital

Referral Pattern for Fetal Echocardiography Over 10 Years in a Single Fetal Imaging Centre from Southern India

Abstract Aim To study the trend in referral patterns for fetal echocardiography (FE) and the outcome Methods Retrospective study of fetuses referred to a single fetal imaging center for FE from Jan 2008 to Dec 2017. The study group was divided into group 1 (2008–2012) and group 2 (2013–2017). Indications were categorized into ‘low-risk’, ‘maternal-risk’ and ‘fetal-risk’. Detection of cardiac defect (CHD) was noted as abnormal outcome and it was analysed in relation to the referral indication. Results The study group had 32,679 cases, 11,468 in group 1 and 21,211 in group 2. Total number of referrals showed an increase of 84% between first and second half of study. ‘Low-risk’ referrals were the most common in both groups but high risk referrals had shown an increase of 24.5% over years. Maternal diabetes was the most common ‘maternal-risk’ factor and abnormal cardiac finding in obstetric scan was the common ‘fetal-risk’ indication. Incidence of CHD increased from 4.6 to 10.2% during the study period. CHD was seen more in ‘fetal-risk’ indications (65%) compared to other risk groups (p < 0.05). Maternal-risk indications individually did not show a significant relationship to CHD. Abnormal cardiac findings in scan, extra-cardiac anomaly, aneuploidy screen positivity, single umbilical artery, growth and liquor abnormality showed a higher risk for CHD. Among low-risk 22.5% showed CHD. Conclusion Referral of high-risk cases for FE has increased recently. Abnormal fetal findings in scan or screening tests was significantly associated with fetal CHD. A significant percentage of CHD was detected in the low-risk group. Strengthening fetal cardiac screening is needed to optimise appropriate risk stratification as well as to increase detection of cardiac anomalies.

Can Ultrasound in Early Gestation Improve Visualization of Fetal Cardiac Structures in Obese Pregnant Women?

(Abstracted from J Ultrasound Med 2019;38:2057–2063) The worldwide obesity epidemic has led to more than 1.9 billion adults 18 years of age or older classified as “obese.” As obesity is associated with greater risk of maternal and feto-neonatal complications, this study aimed to assess the utility of early gestational ultrasounds (USs) in cardiac screening for obese patients. This entailed performing both late first-trimester/early second-trimester transvaginal and transabdominal US, in comparison with the current national guideline-based second-trimester abdominal US for initial fetal cardiac screening in obese patients.

AIUM Practice Parameter for the Performance of Fetal Echocardiography.

AIUM Practice Parameter for the Performance of Fetal Echocardiography.

P08.11: Fetal cardiac screening program in our centre: how to use Doppler imaging at routine screening? A retrospective study

P08.11: Fetal cardiac screening program in our centre: how to use Doppler imaging at routine screening? A retrospective study

P08.01: A novel deep learning based system for fetal cardiac screening

P08.01: A novel deep learning based system for fetal cardiac screening

Usefulness of color Doppler mode for fetal cardiac ultrasound screening in the second trimester: a study at a single perinatal center

Objectives: To evaluate the usefulness of color Doppler in fetal cardiac ultrasound screening in the second trimester.Methods: Fetuses who underwent ultrasound screening at 18–20 weeks’ gestation at Showa University Hospital between 2011 and 2016 were evaluated. After delivery, neonatal congenital heart abnormalities were reviewed and compared with the antenatal ultrasound findings. Since 2014, we have added color Doppler to the routine B mode evaluation of the fetal heart. Congenital heart diseases (CHDs) found antenatally and postnatally were compared before and after protocol alternation. Medical records of all fetuses who underwent ultrasound screening at 18–20 weeks’ gestation at Showa University Hospital between 2011 and 2016 were retrospectively reviewed.Results: There were 47 cases of CHDs confirmed postnatally. The detection rates of CHDs were 45.0% (9/20) in 2011–2013 and 55.6% (15/27) in 2014–2016. In 2011–2013, cases with antenatal diagnosis showed obvious abnormal findings of three-vessel view and four-chamber view with the B mode. In 2014–2016, the detection rate of isolated ventricular septal defect (VSD) was elevated from 10 to 42.9% using color Doppler.Conclusions: In this study, color Doppler improved the detection rate of CHDs. Color Doppler could give us additional information on blood flow although the canal or vessel is too small to detect morphological changes in the second trimester. It might be a useful tool for screening of CHDs with stenosis, regurgitation, and shunt that are difficult to detect by only the B mode in the second trimester.

Can Ultrasound in Early Gestation Improve Visualization of Fetal Cardiac Structures in Obese Pregnant Women?

To determine whether ultrasound (US) imaging of obese pregnant women in early gestation improves fetal cardiac visualization. We report a prospective longitudinal trial comparing late first- and early second-trimester US to traditional second-trimester US for fetal cardiac screening in maternal obesity. All study participants received US for fetal cardiac screening at a gestational age of 13 weeks to 15 weeks 6 days using a transvaginal or transabdominal approach (UST1). All patients also underwent a traditional transabdominal examination at 20 to 22 weeks (UST2). If UST2 failed to complete the cardiac visualization, a follow up examination (second UST2) was performed 2 to 4 weeks later. Imaging was considered complete if the visceral situs, 4 chambers, outflow tracts, and 3 vessels were visualized. The completion rates between UST1 and UST2 were the primary outcomes. A power analysis (85%) and sample size calculation (n = 145) were performed. A total of 152 pregnancies met study criteria. Completion rates of cardiac screening were 77% for UST1 and 70% for UST2 (P = .143). This comparison reached statistical significance in patients with a body mass index of greater than 40 kg/m2 (UST1 [69%] versus UST2 [38%]; P = .001). Sixty-two percent of patients with a body mass index of greater than 40 kg/m2 required second UST2. The highest fetal cardiac screening completion rate was achieved when UST1 was combined with UST2 (96.1%). In 1 fetus, congenital heart disease was detected at UST1 and later confirmed. This study represents the largest prospective trial assessing early-gestation US for fetal cardiac screening in the setting of maternal obesity. In obese pregnancies, the addition of early-gestation US to the 20- to 22-week US substantially improves the visualization of fetal cardiac anatomy.

EP10.24: Usefulness of fetal cardiac ultrasound screening with colour Doppler mode in the second trimester: a study at a single perinatal centre

EP10.24: Usefulness of fetal cardiac ultrasound screening with colour Doppler mode in the second trimester: a study at a single perinatal centre

The effectiveness of extended fetal echocardiography in evaluating fetal cardiac morphology

Purpose: To investigate the feasibility of fetal cardiac screening during routine 11-14 week scan. Materials and Methods: First and second trimester fetal cardiac scan was performed prospectively by sequential segmental analysis methodology in 114 fetuses within an unselected population. Fetal cardiac images were obtained by transabdominal or transvaginal approach at 11+6 – 15+2 gestational weeks at which CRL was between 52.1-98.8 mm and sections of four quadrants, three-vessel trachea, ascending aorta and aortic arch were visualized. Examination time was limited to 15 minutes. First trimester ultrasonographic views were also evaluated by another specialist in order to compare with second trimester cardiac scan and postpartum results. Results: Complete visualization rates were 100% (2/2), 50% (20/40), 35% (21/60), 55% (6/11), and 100% (1/1) for 11, 12, 13, 14, and 15 weeks, respectively. It was found at 44% for all gestational weeks. When aortic arch view excluded complete visualization rate, it was calculated as 74%. Satisfactory visualization was possible in 94.7% (108/114), 89.4% (102/114), 89.4% (102/114), and 58.7% (67/114) of fetuses for four chamber, three-vessels and trachea, ascending aorta and aortic arch, respectively. There was no pathologic findings in the first trimester scanning, which was concordant with second trimester and postpartum period. Conclusions: It is possible to obtain standard echocardiographic images and to perform fetal cardiac scan during first trimester screening. It can be recommended as a screening method for high and low risk patients in experienced hands; however, the scan must definitely be repeated at second trimester in order to detect later lesions by keeping developmental course of many cardiac defects in mind.

Preserving the border and curvature of fetal heart chambers through TDyWT perspective geometry wrap segmentation

Congenital heart defect leads to the structural abnormality in neonatal. Congenital heart defect (CHD) is the major cause for 20% perinatal mortality and 50% infant mortality. However, the fetal cardiac screening plays a vital role to diagnose the CHD during second –trimester. Evaluation of fetal cardiac chamber structure is difficult due to small in size and movement. Four-chamber view has become the first and foremost view of echocardiography to detect structural malformations in the fetal heart. Furthermore, trained obstetricians, Pediatric cardiologists, maternal-fetal medicine specialists, and radiologists need proper knowledge and skills to identify the chamber structure from echocardiography. In addition, the effective diagnosis of fetal heart four-chamber view consumes more time and needs extremely skilled radiologists owing to their low quality, small signal to noise ratio and rapid movement of fetal heart ultrasound images. Thus, the diagnosis of CHD is the most challenging task. In this paper, we propose the Transverse Dyadic Wavelet Transform (TDyWT) algorithm to preserve the border and curvature of four chambers from 18 to 22 weeks ultrasound fetal heart image. We validate the proposed TDyWT algorithm with normal and abnormal images from Mediscan radiological centre. The performance of the TDyWT algorithm analyses qualitatively and quantitatively to prove border and curvature of the chambers is superior to other conventional methods.

Effectiveness of fetal cardiac screening for congenital heart disease using a combination of the four-chamber view and three-vessel view during the second trimester scan.

We aimed to assess the accuracy and effectiveness of fetal cardiac screening for congenital heart disease (CHD) during the second trimester by general obstetricians in a non-selected population. In this multicenter, prospective cohort study of fetal cardiac screening, four-chamber and three-vessel views were recorded by obstetricians at 18-21 gestational weeks (GW). A total of 3005 fetuses that were scheduled for delivery at our institution were included. Thirty-seven newborns were born with CHD (1.2%). On excluding 23 cases of ventricular septal defects, the prenatal detection rate of CHD was 42.8%. Although six cases (75.0%) of severe structural abnormality were diagnosed prenatally, the prenatal detection rate of valvular abnormalities was 0%. One-point ultrasound screening of the fetal heart using a combination of four-chamber and three-vessel views at 18-21 GW by general obstetricians in a non-selected population may be useful for detecting severe structural abnormalities but not valvular abnormalities. However, this limitation may be improved by conducting another fetal cardiac screening at approximately 30 GW along with the routine use of color Doppler.

Basic heart examination: feasibility study of first-trimester systematic simplified fetal echocardiography.

First-trimester fetal cardiac screening examinations in low-risk populations should not have to meet the specifications required for high-risk populations. Our aim was to evaluate a simplified fetal echocardiographic ('basic heart') examination for early detection of severe congenital heart defects in a low-risk population. This was a first-trimester national 'flash study', performed over a 2-week period. Each observer was requested to perform simplified echocardiography without modifying the time and methods deemed necessary for the routine first-trimester ultrasound examination, in fetuses with crown-rump length between 45 and 84 mm. This basic heart assessment used targeted cross-sections of the four-chamber view (4CV) and of the three vessels and trachea (3VT) view, using color and/or directional power Doppler. All examinations were then reviewed offline and scored for quality by a qualified expert. Sixty observers performed a total of 597 first-trimester ultrasound examinations, each performing an average of 10 (range, 1-26) procedures. Examinations were conducted transabdominally (79%; 472/597), transvaginally (3%; 17/597) or both (18%; 108/597). In 8% (45/597) of cases, the fetal back was anterior, in 18% (108/597) it was on the left side, in 63% (377/597) it was posterior and in 11% (67/597) it was on the right side. It became clear during scoring by the expert that, unlike the Herman quality score for nuchal translucency measurement, it was difficult to assess the quality of these images without taking into account normality of the heart itself. Analysis of scores showed that the 4CV was obtained successfully and was deemed normal in 86% (512/597) of the patients, in 7% (41/597) it was deemed technically infeasible and in 7% (44/597) it was deemed feasible but atypical, which may have been due to the presence of an abnormality or to poor quality of the image. The 3VT view was obtained successfully and was normal in 79% (472/597) of the patients, in 13% (78/597) it was technically infeasible and in 8% (47/597) it was deemed feasible but atypical. Both the 4CV and the 3VT view were obtained successfully and were normal in 73% (435/597) of patients. It is possible for a trained operator to perform simplified fetal echocardiography during the routine first-trimester ultrasound examination in a low-risk population. In most cases, our basic heart examination can be used to reassure parents or identify potential problems to be clarified as early as possible in the second trimester. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Ultrasound Examination of the Fetal Heart.

Congenital heart disease (CHD) occurs in 4-13 per 1000 births in the United States. While many risk factors for CHD have been identified, more than 90% of cases occur in low-risk patients. Guidelines for fetal cardiac screening during the second trimester anatomy ultrasound have been developed by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) in order to improve antenatal detection rates and to standardize the fetal cardiac screening examination. Patients found to be at increased risk of CHD because of risk factors or an abnormal screening examination should be referred for second trimester fetal echocardiography. Recently, 3D and 4D ultrasound techniques are being utilized to enhance detection rates and to better characterize cardiac lesions, and several first trimester ultrasound screening markers have been proposed to identify patients at increased risk of CHD. However, detection rates have not improved significantly due to limitations such as cost, access, and training that are associated with new technologies and screening methods. The most cost effective way to improve detection rates of CHD may be to standardize screening protocols across practices according to established guidelines and to have a low threshold for referral for fetal echocardiography.

Assessment of the Cases of Fetal Cardiac Abnormalities Identified Before 22 Weeks' Gestation

As a result of wide spread use of first and second trimester fetal screenings in Japan, the number of cases with fetal abnormalities pointed out prior to 22 weeks’ gestation has increased. In this study, we assessed the cases of fetal cardiac defects identified before 22 weeks’ gestation in our hospital. Transabdominal ultrasonography was made in 55 patients who were pointed out to have fetal abnormalities through the first or second trimester screening. In 36 of these 55 patients, fetal cardiac abnormalities were found. The recordings of the 36 patients were assessed retrospectively. Mean gestational age at fetal abnormality detection was 19.1±3.6 weeks. The 36 cases consisted of 30 morphological abnormalities (83%) and 6 fetal arrhythmias without abnormal cardiac structures (17%). The abnormal findings at the detection included 23 cases of suspicious fetal morphological abnormalities (64%), 8 cases of fetal arrhythmia (8%) and 8 cases with increased nuchal translucency (NT) (8%). Five patients (14%) in 36 chose termination of the pregnancy. In this study, only 8% of fetal cardiac abnormalities were identified by increased NT. It is thought that fetal echocardiography for patients with increased NT is useful for the early detection of fetal cardiac defects. At this point in Japan, first trimester screening by NT measurement is not very common, but it has gradually been increasing. We think it desirable and appropriate to establish a fetal cardiac screening prior to 22 weeks’ gestation.

EP07.06: Impact of maternal obesity on fetal cardiac screening

EP07.06: Impact of maternal obesity on fetal cardiac screening

Identification of fetal cardiac anatomy and hemodynamics: a novel enhanced screening protocol.

BackgroundPrenatal cardiac screening is of great importance as it contributes to appropriate neonatal management and helps parents to make a decision regarding their pregnancy. The aim of our study was to evaluate the efficiency of a newly proposed screening protocol in the detection of fetal congenital heart disease (CHD).MethodsThis was a prospective study. A total of 52 cases of confirmed CHD fetuses and 248 cases of randomly selected normal fetuses were included in the study. Two sonographers with similar experience performed the cardiac screenings under two different protocols independently. The conventional protocol (Protocol A) paid greater attention to the four-chamber view and the outflow tract views. A 6-month training program was provided to sonographers performing scans under the new protocol (Protocol B), which emphasized systematically evaluating fetal cardiac anatomy and hemodynamics. Color Doppler was mandatory and some ultrasonic signs for special cardiac anomalies were also introduced into this protocol.ResultsProtocol B detected more cardiac anomalies than did Protocol A (96.2 % vs. 61.5 %, P < 0.01). Specifically, Protocol B was superior to Protocol A in detecting cardiac malpositions, abnormal systemic and pulmonary venous connection, right aortic arch, transposition of the great arteries, and congenital corrected transposition of the great arteries. By visualizing flow disturbance and retrograde flow with color Doppler, Protocol B was better than Protocol A in screening valvular associated malformations, such as pulmonary atresia, pulmonary stenosis, tricuspid dysplasia, etc. For the normal fetuses, Protocol B was better than Protocol A in reducing the false-positive detection of septal defects.ConclusionsThe current study introduces an enhanced protocol for fetal cardiac screening, under which the obstetric screening sonographers systematically identify fetal cardiac anatomy and hemodynamics. A short-term training program makes it possible for the screening sonographers to become familiar with the new protocol, and its value has been confirmed due to improvements made in screening efficiency.Electronic supplementary materialThe online version of this article (doi:10.1186/s12884-016-0933-9) contains supplementary material, which is available to authorized users.

Congenital Heart Disease and Maternal Diabetes Mellitus

Introduction: Diabetes mellitus is a relatively common illness that can complicatepregnancy and result in an increased incidence of congenital malformations.Offspring of diabetic mothers suffering from type IDDM have a fivefold incidence of congenital malformations compared to pregnancies in the general healthy population. Specifically, the pattern of congenital heart disease(CHD) encountered among this group, with an emphasis on abnormalities of laterality, looping and conotruncal septation, suggesting that the maternal metabolic state affects cardiogenesis at a very early stage of the developmental period, prior to 7 weeks of gestation. Although many have been written on the effect of diabetes in pregnant women, less is known for the effects of type II DM and gestational diabetes mellitus(GDM) and its role in provoking CHD. Aim: Aim of this paper is to review the literature regarding the types of CHD seen in offspring of mothers suffering fromdifferent typesof diabetes mellitus, maternal types 1 and 2 and gestational and to comment on the incidences and any differencesfound in the types of detected CHD. Method: a systematic literature resurge of the last 15 years was reviewed focusing to produce answers on the aims of the study. Conclusion: basedon the existing evidencehigh frequency of CHD can be found in any type of maternal diabetes mellitus. For this reason, we believe that any type of diabetes present in a pregnancy must be a strong indication undergoing specific special fetal cardiac prenatal screening, aiming to detect possible CHD.