Search

Objectives: Measuring anatomical parameters in fetal heart ultrasound images is crucial for the diagnosis of congenital heart disease (CHD), which is highly dependent on the clinical experience of the sonographer. To address this challenge, we propose an automated segmentation method using the channel-wise knowledge distillation technique., Methods: We design a teacher-student architecture to conduct channel-wise knowledge distillation. ROI-based cropped images and full-size images are used for the teacher and student models, respectively. It allows the student model to have both the fine-grained segmentation capability inherited from the teacher model and the ability to handle full-size test images. A total of 1,300 fetal heart ultrasound images of three-vessel view were collected and annotated by experienced doctors for training, validation, and testing., Results: We use three evaluation protocols to quantitatively evaluate the segmentation accuracy: Intersection over Union (IoU), Pixel Accuracy (PA), and Dice coefficient (Dice). We achieved better results than related methods on all evaluation metrics. In comparison with DeepLabv3+, the proposed method gets more accurate segmentation boundaries and has performance gains of 1.8% on mean IoU (66.8% to 68.6%), 2.2% on mean PA (79.2% to 81.4%), and 1.2% on mean Dice (80.1% to 81.3%)., Conclusions: Our segmentation method could identify the anatomical structure in three-vessel view of fetal heart ultrasound images. Both quantitative and visual analyses show that the proposed method significantly outperforms the related methods in terms of segmentation results., Competing Interests: The authors declare that they have no conflicts of interest., (Copyright © 2022 Qiwen Cai et al.)

Background: Fetal heart evaluation has been proven to have the utility in the differential diagnosis of cardiomyopathies or prediction of perinatal mortality in congenital heart disease. The purpose of this study is to introduce a novel method that can measure the global and regional longitudinal strain of left ventricle in fetuses, described as automated cardiac motion quantitation (aCMQ). Furthermore, to evaluate the feasibility and value of aCMQ., Methods: A cross-sectional study was conducted in 138 fetuses with normal heart structure, then were divided into four groups by gestational age, standard four-chamber view clips were collected, the global and segmental longitudinal peak systolic strain of the left ventricle were measured by aCMQ, interobserver and intraobserver variability was analyzed., Results: The success rate of aCMQ analysis was 90.6% (125/138). aCMQ can obtain more comprehensive data to evaluate fetal cardiac function by on-line analysis, including the left ventricular end-systolic volume (LVESV, 0.64 ± 0.16 ml), the left ventricular end-diastolic volume (LVEDV, 0.85 ± 0.35 ml), strain and so on. There was no significant correlation between the gestational age and the global and segmental longitudinal peak strain of the left ventricle (all p  > .05). The base and middle segmental strain values of left ventricular free wall were similar to those of the interventricular septum. No significant differences were found in base segments and middle segments for left ventricular free wall (LVLW) and the interventricular septum (IVS) ( p  > .05). The strain of the apex segment was higher than that of the base and middle segment ( p  < .05), and there were significant differences between the apex and the base and the middle ( p  < .05). Interobserver and intraobserver variability showed a small bias among the observers., Conclusions: The global and regional strain values of fetus have no correlation with gestational ages, and the strain values of apical segment are higher than those of basal and middle segments. The base and middle segmental strain values of left ventricular free wall were similar to those of the interventricular septum. aCMQ is a novel method of two dimensional speckle tracking echocardiography to assess the global and regional systolic function of fetuses. It's a feasible and reproducible approach to evaluate normal cardiac function of fetus quantitatively and may have potential in fetuses with congenital heart diseases.

Objectives: To explore the feasibility of applying smart-planes fetal heart (S-Planes FH) to 3D volumes to generate and display the outflow tract views (OFTVs) and evaluate its agreement and reliability in determining fetal vessel dimensions., Methods: A total of 147 normal fetuses in the second or third trimester were enrolled. Once conventional 2D sonographic examination was finished, one or more 3D static volumes were then acquired in the 4-chamber view (4CV). The OFTVs automatically generated and displayed via a simple offline operation using S-Planes FH software. Per fetus, a single qualified S-Planes FH image was selected by an expert. The inner diameters of the aorta (AO) and pulmonary artery (PA) were measured in 2D and S-Planes FH images, respectively. The agreement and reliability of the two methods were verified by the Bland-Altman method and the intraclass correlation coefficient (ICC), respectively., Results: A total of 515 3D static volumes were successfully acquired in 147 fetuses. The OFTVs were generated and displayed using S-Planes FH in 138 (93.9%) cases. S-Planes FH had good agreement in measuring the diameters of great arteries (95% CI of limits, -0.9419 to 1.1464 for AO diameter, -0.8871 to 1.7007 for PA diameter). The intraobserver and interobserver ICCs in this study were greater than 0.81, indicating almost perfect reliability of the two methods with regard to observers., Conclusions: S-Planes FH software is a feasible method for generating OFTVs as well as determining vessel dimensions in the second and third trimesters and had good agreement with 2D examination. Therefore, it possesses clinical value in screening for CHD and can be operated by different sonographers.

Objective: To compare the accuracy, efficiency, and consistency between experienced and less-experienced professionals using intelligent navigation echocardiography., Methods: In this prospective study, we enrolled 93 second- and third-trimester fetuses with conotruncal defects (CTD) from July 2017 to February 2018. One or more spatiotemporal image correlation volume data sets were collected per case. The fetuses with CTD were diagnosed by the following two groups of professionals (n = 20 in each) with different experience levels using intelligent navigation echocardiography and two-dimensional ultrasound: group A with 15 years of experience and group B with 1 year of experience. The diagnostic consistency and accuracy of the technologies between the two groups were analyzed., Results: Satisfactory consistency was noted in the two groups (group A, τ = 0.855, P < 0.05, and group B, τ = 0.821, P < 0.05), and no significant difference in accuracy (χ 2  = 3.218, P > 0.05) in using intelligent navigation echocardiography was reported between the two groups. However, there a significant difference in accuracy (χ 2  = 0.021, P < 0.05) when using two-dimensional ultrasound was observed between the two groups., Conclusion: Intelligent navigation echocardiography was found to be efficient and accurate for the diagnosis of CTD and good consistency existed in the experienced and less-experienced professionals., (© 2020 International Federation of Gynecology and Obstetrics.)

The main objective of this study was to investigate the diagnostic performance of FINE in generating and displaying 3 specific abnormal fetal echocardiography views such as left ventricular outflow tract (LVOT) view, right ventricular outflow tract (RVOT) view, and 3-vessels and trachea (3VT) view in fetuses with double-outlet right ventricle (DORV). In this prospective study, thirty fetuses diagnosed with DORV by fetal echocardiography in the second and third trimesters were enrolled. One or more STIC volume data-sets were collected from the 4-chamber view as initial view for each fetus, one optimal volume per fetus was selected for on-line analysis using FINE, and the diagnosis plane image was optimized using the Virtual Intelligent Sonographer Assistance (VIS-assistance).The visualization rates of 3 specific abnormal fetal echocardiography views of DORV and key diagnostic elements were calculated. One or more STIC volumes (n = 30 total) were obtained in 25 patients. A single STIC volume per patient was analyzed using the FINE method. FINE was able to successfully generate and display 3 specific abnormal fetal echocardiography views. The display rates of the 3 specific abnormal fetal echocardiography views (3VT, LVOT, RVOT) were 84.0%, 76.0% and 84.0%, respectively. By applying intelligent navigation technology to STIC volume data-sets, the FINE method can successfully generate three specific abnormal cardiac fetal echocardiography diagnostic views in fetuses with DORV, the FINE method can be used for screening and remote consultation of fetal DORV.

Morphology and function in a fetal heart with severe tricuspid regurgitation remains challenging. The aim of this study was to assess cardiac morphology and function in fetuses with severe tricuspid regurgitation by fetal heart quantification (HQ) and to assess the practical value of fetal HQ. Clinical information was analyzed for 63 pregnant women who underwent fetal cardiac ultrasonography. The women were divided into those who had a fetus with severe tricuspid regurgitation (n = 20) and those with a normal fetus (n = 40). The global sphericity index (GSI), fractional area change (FAC), and global longitudinal strain (GLS) of both ventricles and the sphericity index (SI) and fractional shortening (FS) of 24 segments were quantified by fetal HQ using speckle tracking imaging. Fetuses with severe tricuspid regurgitation had a significantly lower GSI (1.14 ± 0.10 vs. 1.26 ± 0.08, p < 0.001) and a higher GSI Z-score (-0.98 ± 1.01 vs. 0.25 ± 0.87, p < 0.001) as well as a significantly lower right ventricular FAC (36.50 ± 7.34% vs. 45.19 ± 3.39%, p < 0.001), FAC Z-score (-1.02 ± 1.41 vs. 0.49 ± 0.74, p < 0.001), and GLS (-21.01 ± 5.66% vs. 45.19 ± 3.49%, p < 0.001). The SI and SI Z-score were significantly lower in segments 1–18 of the right ventricle in fetuses with severe tricuspid regurgitation (p < 0.05); furthermore, FS of segments 1–12 and 19–24 and the FS Z-score of segments 18–24 were significantly lower in fetuses with severe tricuspid regurgitation (p < 0.05). Fetal HQ is useful for evaluation of cardiac morphology and function in fetuses with severe tricuspid regurgitation and can provide important reference information for both clinical diagnosis and treatment. [ABSTRACT FROM AUTHOR]

Objective: This study investigated the feasibility and accuracy of fetal intelligent navigation echocardiography (FINE) for the prenatal diagnosis of congenital heart disease (CHD) by inexperienced and experienced operators. Method: In this prospective study, all volume data sets from 120 fetuses with a broad spectrum of CHD were acquired using spatiotemporal image correlation technology. The prenatal diagnostic procedures were performed by two operators with different experience (beginner: 1 year and expert: 15 years) using FINE and traditional fetal echocardiography. Data were analyzed on the time of examination and acquisition of results. Results: Diagnoses made by FINE and traditional echocardiography were completely consistent with the final diagnosis of CHD in 98 (81.66%) versus 20 (16.66%) (P < 0.001) beginners and 87.50% (n = 105) versus 101 (84.16%) experts, respectively. On the contrary, there was significant difference using traditional echocardiography (16.66% versus 84.16%, P < 0.001) by two examiners. Furthermore, the examination time decreased when using FINE compared with using traditional echocardiography (beginner operators: 4.54 ± 1.03 min versus 20.58 ± 3.36 min, P < 0.001; expert operators: 3.89 ± 0.96 min versus 12.73 ± 1.62 min, P < 0.001). Conclusion: Based on our results, a prenatal diagnosis of CHD can be made with high feasibility and accuracy using FINE compared with traditional fetal echocardiography for beginner operators. Synopsis: Intelligent navigation echocardiography was effective and accurate for the diagnosis of congenital heart disease compared with traditional echocardiography when performed by less‐experienced operators. [ABSTRACT FROM AUTHOR]