Learning curve and factors influencing the feasibility of performing fetal echocardiography at the time of the first-trimester scan

Learning deep architectures for the interpretation of first-trimester fetal echocardiography (LIFE) - a study protocol for developing an automated intelligent decision support system for early fetal echocardiography

BACKGROUND

Congenital Heart Disease represents the most frequent fetal malformation. The lack of prenatal identification of congenital heart defects can have adverse consequences for the neonate, while a correct prenatal diagnosis of specific cardiac anomalies improves neonatal care neurologic and surgery outcomes. Sonographers perform prenatal diagnosis manually during the first or second-trimester scan, but the reported detection rates are low. This project's primary objective is to develop an Intelligent Decision Support System that uses two-dimensional video files of cardiac sweeps obtained during the standard first-trimester fetal echocardiography (FE) to signal the presence/absence of previously learned key features.

METHODS

The cross-sectional study will be divided into a training part of the machine learning approaches and the testing phase on previously unseen frames and eventually on actual video scans. Pregnant women in their 12-13 + 6 weeks of gestation admitted for routine first-trimester anomaly scan will be consecutively included in a two-year study, depending on the availability of the experienced sonographers in early fetal cardiac imaging involved in this research. The Data Science / IT department (DSIT) will process the key planes identified by the sonographers in the two- dimensional heart cine loop sweeps: four-chamber view, left and right ventricular outflow tracts, three vessels, and trachea view. The frames will be grouped into the classes representing the plane views, and then different state-of-the- art deep-learning (DL) pre-trained algorithms will be tested on the data set. The sonographers will validate all the intermediary findings at the frame level and the meaningfulness of the video labeling.

DISCUSSION

FE is feasible and efficient during the first trimester. Still, the continuous training process is impaired by the lack of specialists or their limited availability. Therefore, in our study design, the sonographer benefits from a second opinion provided by the developed software, which may be very helpful, especially if a more experienced colleague is unavailable. In addition, the software may be implemented on the ultrasound device so that the process could take place during the live examination.

TRIAL REGISTRATION

The study is registered under the name "Learning deep architectures for the Interpretation of Fetal Echocardiography (LIFE)", project number 408PED/2020, project code PN-III-P2-2.1-PED-2019.

TRIAL REGISTRATION

ClinicalTrials.gov , unique identifying number NCT05090306, date of registration 30.10.2020.

First Trimester Tricuspid Regurgitation: Clinical Significance

Tricuspid regurgitation is a cardiac valvular anomaly that consists of the return of blood to the right atrium during systole due to incomplete valve closure. This structure can be visualized on ultrasound between 11 and 14 weeks of gestation in most cases. Despite being a common finding, even in healthy fetuses, the presence of tricuspid regurgitation may be associated with chromosomal and structural abnormalities. The evaluation of tricuspid flow and the presence of regurgitation on first-trimester ultrasound has shown promising results regarding its role in the early detection of aneuploidies, congenital heart defects, and other adverse perinatal outcomes. This review article aims to demonstrate the importance of tricuspid regurgitation as a secondary marker, and consequently, significant benefits of its early detection when added to the combined first-trimester screening. Its value will be discussed, namely its sensitivity and specificity, alone and together with other current markers in the fetal assessment performed in the first-trimester ultrasound.

HDlive Flow Silhouette with spatiotemporal image correlation for assessment of fetal cardiac structures at 12 to 14 + 6 weeks of gestation

OBJECTIVES

To assess fetal cardiac structures using HDlive Flow Silhouette with spatiotemporal image correlation (STIC) at 12 to 14 + 6 weeks of gestation, and verify the feasibility of obtaining five cardiac views in the late first and early second trimesters of pregnancy. The fetal cardiac shape and the aspect of the descending aorta were also evaluated.

METHODS

Eighty normal fetuses at 12 to 14 + 6 weeks of gestation were studied using trans-abdominal HDlive Flow Silhouette with STIC to assess the feasibility of five fetal cardiac views (frontal, spatial three-vessel, panoramic, posterior, and right lateral views). Target structures in each cardiac view were evaluated. 'Good' was assigned when all structures were noted, 'Fair' when only one structure was missed, and 'Poor' when two and more structures could not be detected. Frequencies of an elongated heart and those of a tortuous descending aorta were counted.

RESULTS

Forty-nine fetuses were effectively included in the analysis. Success rates of 'Good' and 'Fair' were significantly higher with spatial three-vessel (p<0.01) and panoramic views (p<0.05). Frequencies of "Elongated heart", "Elongated left ventricle", and "Spherical heart" were 12.2, 6.2, and 81.6%, respectively. Frequencies of "Tortuous descending aorta" and "Straight descending aorta" were 12.2 and 87.8%, respectively.

CONCLUSIONS

The feasibility of obtaining fetal five cardiac views using HDlive Flow Silhouette with STIC is good, and this technique provides useful information for evaluating fetal cardiac structures in the late first and early second trimesters of pregnancy.

First-Trimester Fetal Echocardiography: Identification of Cardiac Structures for Screening from 6 to 13 Weeks' Gestational Age

BACKGROUND

Early fetal echocardiography (FE), performed at 12 to 16 weeks' gestational age (GA), can be used to screen for fetal heart disease akin to that routinely performed in the second trimester. The efficacy of FE at earlier GAs has not been as well explored, particularly with recent advances in ultrasound technology. The aim of this study was to evaluate the efficacy of early FE in assessing fetal heart structure, and the added benefit of color Doppler (CD), from as early as 6 weeks through to 13⁺⁶ weeks' GA.

METHODS

Pregnant women were prospectively recruited for first-trimester FE. All underwent two-dimensional (2D) cardiac imaging combined with CD assessment, and all were offered second-trimester fetal echocardiographic evaluations. Fetal cardiac anatomy was assessed both in real time during FE and additionally offline by two separate reviewers.

RESULTS

Very early FE was performed in 202 pregnancies including a total of 261 fetuses, with 92% (n = 241) being reassessed at ≥18 weeks' GA. Mean GA at FE was 10⁺⁶ weeks (range, 6⁺¹ to 13⁺⁶ weeks). Transabdominal scanning was used in all cases, and transvaginal scanning was used additionally in most at <11 weeks' GA (n = 103 of 117 [88%]). There was stepwise improvement in image resolution of the fetal heart in those pregnancies that presented at later gestation for assessment. CD assisted with definition of cardiac anatomy at all GAs. A four-chambered heart could be identified in 52% of patients in the eighth week (n = 12 of 23), improving to 80% (n = 36 of 45) in the 10th week and 98% (n = 57 of 58) by the 11th week. The inferior vena cava was visualized by 2D imaging in only 4% (n = 1 of 23) in the eighth week, increasing to 13% (n = 6 of 45) by the 10th week and 80% (n = 25 of 31) by the 13th week. CD improved visualization of the inferior vena cava at earlier GAs to >80% (n = 37 of 45) from 10 weeks. Pulmonary veins were not visualized by either 2D imaging or CD until after the 11th week. Both cardiac outflow tracts could be visualized by 2D imaging in the minority from 8⁺⁰ to 10⁺⁶ weeks (n = 18 of 109 [16%]) but were imaged in most from 11⁺⁰ to 13⁺⁶ weeks (n = 114 of 144 [79%]). CD imaging improved visualization of both outflow tracts to 64% (n = 29 of 45) in the 10th week. On 2D imaging alone, both the aortic and ductal arches were seen in only 29% of patients in the 10th week (n = 13 of 45), increasing to 58% when CD was used (58% [n = 26 of 45]) and to >80% (n = 47 of 58) using CD in the 11th week.

CONCLUSIONS

Very early FE, from as early as 8 weeks, can be used to assess cardiac structures. The ability to image fetal heart structures between 6 and 8 weeks is currently nondiagnostic. The use of CD significantly increases the detection of cardiac structures on early FE. The ideal timing of complete early FE, excluding pulmonary vein assessment, appears to be after 11 weeks' GA.

Advances in ultrasound imaging for congenital malformations during early gestation

BACKGROUND

With refinement in ultrasound technology, detection of fetal structural abnormalities has improved and there have been detailed reports of the natural history and expected outcomes for many anomalies. The ability to either reassure a high-risk woman with normal intrauterine images or offer comprehensive counseling and offer options in cases of strongly suspected lethal or major malformations has shifted prenatal diagnoses to the earliest possible gestational age.

METHODS

When indicated, scans in early gestation are valuable in accurate gestational dating. Stricter sonographic criteria for early nonviability guard against unnecessary intervention. Most birth defects are without known risk factors, and detection of certain malformations is possible in the late first trimester.

RESULTS

The best time for a standard complete fetal and placental scan is 18 to 20 weeks. In addition, certain soft anatomic markers provide clues to chromosomal aneuploidy risk. Maternal obesity and multifetal pregnancies are now more common and further limit early gestation visibility.

CONCLUSION

Other advanced imaging techniques during early gestation in select cases of suspected malformations include fetal echocardiography and magnetic resonance imaging.

Acoustic output measured by thermal and mechanical indices during fetal echocardiography at the time of the first trimester scan

We measured acoustic output, expressed as the thermal index (TI) and mechanical index (MI), during fetal echocardiography at the time of the first trimester scan. TI and MI were retrieved from the saved displays during gray-mode, high-definition color flow Doppler and pulsed-wave Doppler (tricuspid flow) ultrasound examinations of the fetal heart and from the ductus venosus assessment. A total of 399 fetal cardiac examinations were evaluated. There was a significant increase in TI values from B-mode studies (0.07 ± 0.04 [mean ± SD]) to color flow mapping (0.2 ± 0.0) and pulsed-wave Doppler studies (0.36 ± 0.05). The TI from ductus venosus assessment (0.1 ± 0.01) was significantly lower than those from Doppler examinations of the heart. MI values from B-mode scans (0.65 ± 0.12) and color flow mapping (0.71 ± 0.11) were comparable, although different, and both values were higher than those from pulsed-wave Doppler tricuspid evaluation (0.39 ± 0.03). There were no differences in MI values from power Doppler assessment between the tricuspid flow and ductus venosus. Safety indices were remarkably stable and were largely constant, especially for color Doppler (TI), tricuspid flow (MI) and ductus venosus assessment (TI, MI). We acquired satisfactory Doppler images and/or signals at acoustic levels that were lower than the actual recommendations and never reached a TI of 0.5.

Factors affecting the feasibility of routine first-trimester fetal echocardiography

OBJECTIVES

The purpose of our study was to assess the factors that may improve the feasibility of routine fetal echocardiography at the time of the first-trimester scan.

METHODS

In this prospective study, we included 616 healthy singleton fetuses. Examinations were performed transabdominally by a single sonographer at the beginning of his training in first-trimester screening. The fetal heart was examined by high-definition color Doppler imaging to obtain the 4-chamber view, right and left ventricular outflow tracts, and 3-vessel and trachea view. Logistic regression was used to investigate the effect on the ability to visualize different cardiac structures.

RESULTS

The frequency of successful heart examinations increased significantly with the number of scans performed (P < .05). The sonographer needed 180 examinations before he could successfully examine the heart in at least 80% of cases. Significant factors that increased the probability of adequate echocardiography were the length of the heart examination and the experience of the sonographer (P< .05) but not transducer-heart distance, maternal body mass index, fetal crown-rump length, placenta interposition, or restrictive fetal position. Visualization of the left ventricular outflow tract could be improved by increasing the experience of the sonographer and decreasing the transducer-heart distance. Also, visualization of the 3-vessel and trachea view depended on the length of the heart examination, the experience of the sonographer, an anterior position of the placenta, and a restrictive fetal position.

CONCLUSIONS

Competence in color flow mapping assessment of the fetal heart at gestational ages of 11 weeks to 13 weeks 6 days is achieved only after extensive supervised training.

Fetal heart assessment in the first trimester of pregnancy: influence of crown-rump length and maternal body mass index

OBJECTIVE

To evaluate the influence of the crown-rump length and body mass index on sonographic evaluation of the fetal heart using abdominal and vaginal routes in the first trimester of pregnancy.

METHODS

We conducted a cross-sectional study with 57 pregnant women between 12-14 weeks (CRL< 84 mm). We evaluated the following fetal cardiac plans using the abdominal and vaginal routes: four-chamber view, right ventricle outflow tract, left ventricle outflow tract and aortic arch. We used the B-mode, color Doppler and four-dimensional ultrasonography (spatio-temporal image correlation). To evaluate the influence of crown-rump length and body mass index in the assessment of fetal cardiac planes, we used the t test unpaired.

RESULTS

There were no statistically significant differences in the rates of success and failure between abdominal and vaginal routes in relation to body mass index, however, there was a higher failure rate in vaginal assessment using B mode associated with color Doppler (P<0.01).

CONCLUSION

The crown-rump length and body mass index had no interference in fetal cardiac assessment in the first trimester of pregnancy.