Intracardiac Blood Flow Velocities Research Articles

Deep learning-based prediction of intra-cardiac blood flow in long-axis cine magnetic resonance imaging

Purpose: We aimed to design and evaluate a deep learning-based method to automatically predict the time-varying in-plane blood flow velocity within the cardiac cavities in long-axis cine MRI, validated against 4D flow. Methods: A convolutional neural network (CNN) was implemented, taking cine MRI as the input and the in-plane velocity derived from the 4D flow acquisition as the ground truth. The method was evaluated using velocity vector end-point error (EPE) and angle error. Additionally, the E/A ratio and diastolic function classification derived from the predicted velocities were compared to those derived from 4D flow. Results: For intra-cardiac pixels with a velocity > 5 cm/s, our method achieved an EPE of 8.65 cm/s and angle error of 41.27°. For pixels with a velocity > 25 cm/s, the angle error significantly degraded to 19.26°. Although the averaged blood flow velocity prediction was under-estimated by 26.69%, the high correlation (PCC = 0.95) of global time-varying velocity and the visual evaluation demonstrate a good agreement between our prediction and 4D flow data. The E/A ratio was derived with minimal bias, but with considerable mean absolute error of 0.39 and wide limits of agreement. The diastolic function classification showed a high accuracy of 86.9%. Conclusion: Using a deep learning-based algorithm, intra-cardiac blood flow velocities can be predicted from long-axis cine MRI with high correlation with 4D flow derived velocities. Visualization of the derived velocities provides adjunct functional information and may potentially be used to derive the E/A ratio from conventional CMR exams.

Imaging and modeling of blood flow in the developing heart

Blood flow during early embryonic development modulates the growth and remodeling (volume and shape) of the heart. Several studies in animals have shown that altered hemodynamic conditions during embryonic development resulted in cardiac defects resembling those in the 1% of human newborns with congenital heart disease. To better understand the relationship between blood flow and cardiac development, as well as the mechanisms by which blood flow affects heart growth, we must quantify intracardiac blood-flow velocities during early embryonic development. There are, however, two main challenges in measuring blood-flow dynamics in the developing heart. The heart is tiny (<1mm) and it beats fast ( 2Hz). Both limit the accuracy of obtaining blood-flow velocities. To overcome that difficulty, we combined computational modeling and imaging in vivo. We used a chicken embryo and optical coherence tomography (OCT) to image the beating, developing heart. During early development, the hearts of humans and chickens are very similar. The chick embryo and its heart are easy to access after opening the egg shell. OCT is a noninvasive optical-imaging technique that allows us to look at tissue sections in vivo with micrometer precision. Moreover, our OCT system simultaneously acquires structural (morphological) andDoppler (velocity) images, which allow measurements of both cardiac geometry and blood-flow velocity during the beating cycle (see Figure 1). Images of the outflow tract (OFT) of the developing chick heart at Hamburger-Hamilton (HH) stage 18 ( 3 out of 21 days of incubation) reveal a complex interaction between blood-flow dynamics and the heart’s beating motion.1, 2 To handle the imaging-speed limitations of our OCT system in picturing rapidly changing 3D cardiac geometry over the beating cycle, we developed a synchronization procedure that allows reconstruction of a 4D image of the heart (3D cardiac Figure 1. Simultaneous structural (top) and Doppler (bottom) optical coherence tomography (OCT) images of the cardiac outflow tract (OFT) of a chick embryo showing (A) lateral and (B) cross-sectional views when the OFT walls are expanded.

First‐trimester fetuses with increased nuchal translucency do not show altered intracardiac flow velocities

To study intracardiac flow velocities in first-trimester fetuses with normal nuchal translucency thickness (NT) and those with increased NT. Ultrasound examinations were performed in 85 normal fetuses and 45 fetuses with NT > 95(th) percentile. Follow-up was complete and postmortem examination was performed on terminated pregnancies. Flow velocities during the early (e-wave) and late (a-wave) peaks across the tricuspid and mitral valves were measured and compared, using multilevel analysis, between the fetuses with normal and those with increased NT. In the group with increased NT, fetuses with and without a heart defect irrespective of the karyotype were compared, and in this group, euploid and aneuploid fetuses were compared, irrespective of the presence of a heart defect. No difference in intracardiac flow velocities was found between fetuses with normal and those with increased NT. Within the group of fetuses with increased NT, there was no difference between the fetuses with and without a cardiac defect. However, comparison of aneuploid with euploid fetuses within the group with increased NT showed that both the e-wave and a-wave peaks were decreased significantly by 3.03 cm/s and 5.95 cm/s, respectively, across the tricuspid valve, and by 3.47 cm/s and 5.92 cm/s, respectively, across the mitral valve (P < 0.05). The most common cardiac malformations were septal defects. There is no difference in intracardiac blood flow velocities between normal fetuses and those with increased NT. This contradicts the theory that NT is caused by impaired atrial contraction or cardiac failure. In fetuses with increased NT, those with aneuploidy show a decreased e-wave and a-wave compared with euploid fetuses. This cannot, however, be explained by the presence of cardiac defects, because there is no difference between fetuses with and without a cardiac defect. Therefore, we hypothesize that the relationship between enlarged NT and cardiac defects can only be explained by a developmental process that coexists at this period of gestation and is linked to cardiovascular development.

Three-dimensional color Doppler flow reconstruction and its clinical applications.

The visualization and quantification of intracardiac blood flow have always been a challenging task for the cardiologist. The advent of color Doppler flow imaging substantially enhanced the clinical diagnosis of heart valve disease. Three-dimensional (3-D) color Doppler, a new diagnostic procedure, refines the diagnostic value of color Doppler by providing unique spatial and temporal information about the actual extension, direction, origin, and size of intracardiac flows. Here, we describe the procedure for 3-D color Doppler reconstruction of intracardiac blood flow velocities and reveal the varied findings in different heart pathologies that cause blood flow disturbances. An automated procedure for the segmentation of turbulent and laminar flows, which allows for the measurement of mitral regurgitant jet volumes, is one of the first 3-D quantitative approaches to the clinical assessment of mitral valve regurgitation. The major technical advances of this procedure include the direct use of digital color Doppler velocity data and an automatic voxel count of the turbulent jet flows. Three-dimensional color Doppler not only can disclose the spatial complex geometry of intracardiac blood flow disturbances but also can quantitatively assess the severity of mitral valve regurgitation.

Quantitative relation between increased intrapericardial pressure and Doppler flow velocities during experimental cardiac tamponade

To establish whether a quantitative relation exists between pericardial pressure and respiratory variation in intracardiac blood flow velocities, a spontaneously breathing closed chest canine model of pericardial tamponade was created. In seven dogs, pericardial pressure was sequentially increased in stages from a mean of -4 +/- 1 to 10 +/- 2 mm Hg while aortic and pulmonary Doppler flow velocities, pleural pressure changes (respiratory effort), blood pressure and cardiac output were measured. The variation in the Doppler-detected peak transaortic velocity (AV) during inspiration (IV) increased linearly from -5 +/- 3% at baseline (pericardial pressure -4 mm Hg) to -32 +/- 9% at a pericardial pressure of 10 mm Hg [IVAV = -2 (pericardial pressure)--13.1; r = 0.78, p less than 10(-6)]. The inspiratory variation in the peak transpulmonary velocity increased from 13 +/- 3% at baseline to 71 +/- 19% at a pericardial pressure of 10 mm Hg. The inspiratory variation in the pulmonary Doppler peak velocity (IVPV) was dependent on both pericardial pressure and degree of respiratory effort [IVPV = 3.8 (pericardial pressure) + 2.6 (respiratory effort) + 10.9; r = 0.88, p less than 10(-8)]. Thus, quantitative relations exist between increases in intrapericardial pressure and increases in inspiratory variation of peak aortic and pulmonary flow velocities. Additionally, pulmonary artery flow velocity is influenced more than aortic velocity by intrathoracic pressure.