Two-Dimensional Echocardiography Estimates of Fetal Ventricular Mass throughout Gestation.

BACKGROUND: Two-dimensional (2D) ultrasound quality has improved in recent years. Quantification of cardiac dimensions is important to screen and monitor certain fetal conditions. We assessed the feasibility and reproducibility of fetal ventricular measures using 2D echocardiography, reported normal ranges in our cohort, and compared estimates to other modalities. METHODS: Mass and end-diastolic volume were estimated by manual contouring in the four-chamber view using TomTec Image Arena 4.6 in end diastole. Nomograms were created from smoothed centiles of measures, constructed using fractional polynomials after log transformation. The results were compared to those of previous studies using other modalities. RESULTS: A total of 294 scans from 146 fetuses from 15+0 to 41+6 weeks of gestation were included. Seven percent of scans were unanalysable and intraobserver variability was good (intraclass correlation coefficients for left and right ventricular mass 0.97 [0.87-0.99] and 0.99 [0.95-1.0], respectively). Mass and volume increased exponentially, showing good agreement with 3D mass estimates up to 28 weeks of gestation, after which our measurements were in better agreement with neonatal cardiac magnetic resonance imaging. There was good agreement with 4D volume estimates for the left ventricle. CONCLUSION: Current state-of-the-art 2D echocardiography platforms provide accurate, feasible, and reproducible fetal ventricular measures across gestation, and in certain circumstances may be the modality of choice.

Fully Automated 3-D Ultrasound Segmentation of the Placenta, Amniotic Fluid, and Fetus for Early Pregnancy Assessment.

Volumetric placental measurement using 3-D ultrasound has proven clinical utility in predicting adverse pregnancy outcomes. However, this metric cannot currently be employed as part of a screening test due to a lack of robust and real-time segmentation tools. We present a multiclass (MC) convolutional neural network (CNN) developed to segment the placenta, amniotic fluid, and fetus. The ground-truth data set consisted of 2093 labeled placental volumes augmented by 300 volumes with placenta, amniotic fluid, and fetus annotated. A two-pathway, hybrid (HB) model using transfer learning, a modified loss function, and exponential average weighting was developed and demonstrated the best performance for placental segmentation (PS), achieving a Dice similarity coefficient (DSC) of 0.84- and 0.38-mm average Hausdorff distances (HDAV). The use of a dual-pathway architecture improved the PS by 0.03 DSC and reduced HDAV by 0.27 mm compared with a naïve MC model. The incorporation of exponential weighting produced a further small improvement in DSC by 0.01 and a reduction of HDAV by 0.44 mm. Per volume inference using the FCNN took 7-8 s. This method should enable clinically relevant morphometric measurements (such as volume and total surface area) to be automatically generated for the placenta, amniotic fluid, and fetus. The ready availability of such metrics makes a population-based screening test for adverse pregnancy outcomes possible.

Fully automated, real-time 3D ultrasound segmentation to estimate first trimester placental volume using deep learning.

We present a new technique to fully automate the segmentation of an organ from 3D ultrasound (3D-US) volumes, using the placenta as the target organ. Image analysis tools to estimate organ volume do exist but are too time consuming and operator dependant. Fully automating the segmentation process would potentially allow the use of placental volume to screen for increased risk of pregnancy complications. The placenta was segmented from 2,393 first trimester 3D-US volumes using a semiautomated technique. This was quality controlled by three operators to produce the "ground-truth" data set. A fully convolutional neural network (OxNNet) was trained using this ground-truth data set to automatically segment the placenta. OxNNet delivered state-of-the-art automatic segmentation. The effect of training set size on the performance of OxNNet demonstrated the need for large data sets. The clinical utility of placental volume was tested by looking at predictions of small-for-gestational-age babies at term. The receiver-operating characteristics curves demonstrated almost identical results between OxNNet and the ground-truth). Our results demonstrated good similarity to the ground-truth and almost identical clinical results for the prediction of SGA.