Identification of hemodynamic biomarkers for bicuspid aortic valve induced aortic dilation using machine learning

Changes in aorta hemodynamics in Left-Right Type 1 bicuspid aortic valve patients after replacement with bioprosthetic valves: An in-silico study

Bicuspid aortic valve (BAV) is the most common cardiac congenital abnormality with a high rate of concomitant aortic valve and ascending aorta (AAo) pathologic changes throughout the patient's lifetime. The etiology of BAV-related aortopathy was historically believed to be genetic. However, recent studies theorize that adverse hemodynamics secondary to BAVs also contribute to aortopathy, but their precise role, specifically, that of wall shear stress (WSS) magnitude and directionality remains controversial. Moreover, the primary therapeutic option for BAV patients is aortic valve replacement (AVR), but the role of improved post-AVR hemodynamics on aortopathy progression is also not well-understood. To address these issues, this study employs a computational fluid dynamics model to simulate personalized AAo hemodynamics before and after TAVR for a small cohort of 6 Left-Right fused BAV patients. Regional distributions of five hemodynamic metrics, namely, time-averaged wall shear stress (TAWSS) and oscillating shear index (OSI), divergence of wall shear (DWSS), helicity flux integral & endothelial cell activation potential (ECAP), which are hypothesized to be associated with potential aortic injury are computed in the root, proximal and distal ascending aorta. BAVs are characterized by strong, eccentric jets, with peak velocities exceeding 4 m/s and axially circulating flow away from the jets. Such conditions result in focused WSS loading along jet attachment regions on the lumen boundary and weaker, oscillating WSS on other regions. The jet attachment regions also show alternating streaks of positive and negative DWSS, which may increase risk for local tissue stretching. Large WSS magnitudes, strong helical flows and circumferential WSS have been previously implicated in the progression of BAV aortopathy. Post-intervention hemodynamics exhibit weaker, less eccentric jets. Significant reductions are observed in flow helicity, TAWSS and DWSS in localized regions of the proximal AAo. On the other hand, OSI increases post-intervention and ECAP is observed to be low in both pre- and post-intervention scenarios, although significant increases are also observed in this ECAP. These results indicate a significant alleviation of pathological hemodynamics post AVR.

Enhancement of intra-cardiac flow-field data using adaptive Kernel filtering

A method of determining the optimal kernel size for filtering noise in vortex dominated flow-fields, as found in the cardiac chambers is presented in this paper. Using synthetic flow fields generated using harmonic functions and perturbed using Gaussian noises of different amplitudes and spreads, the effect of kernel size on noise removal using the Median filter is tested systematically. It is shown that there exists an optimal kernel size at which the Median filter works best. The size of the optimal kernel is shown to be related to the vortex size. When applied to MRI generated cardiac flow-fields, the approach is seen to reveal underlying vortex patterns thereby aiding as an effective tool in the diagnosis and prognosis of cardiac diseases based on vortices as clinical biomarkers. The behavior of the restored cardiac flow fields which are filtered with the optimal kernel size and also with some values preceding and succeeding it are similar to that observed in studies with synthetic flow fields. This confirms that the optimal size of the kernel is related to the cardiac vortex size as is observed in the case of synthetic flow fields.

Multiscale model for blood flow after a bileaflet artificial aortic valve implantation

Cardiovascular diseases are the leading cause of mortality in the world, mainly due to atherosclerosis and its consequences. The article presents the numerical model of the blood flow through artificial aortic valve. The overset mesh approach was applied to simulate the valve leaflets motion and to realize the moving mesh, in the aortic arch and the main branches of cardiovascular system. To capture the cardiac system's response and the effect of vessel compliance on the outlet pressure, the lumped parameter model has been also included within the solution procedure. Three different turbulence modeling approaches were used and compared - the laminar, k-ϵ and k-ω model. The simulation results were also compared with the model excluding the moving valve geometry and the importance of the lumped parameter model for the outlet boundary condition was analyzed. Proposed numerical model and protocol was found as suitable for performing the virtual operations on the real patient vasculature geometry. The time-efficient turbulence model and overall solving procedure allows to support the clinicians in making decisions about the patient treatment and to predict the results of the future surgery.

Advances in machine learning applications for cardiovascular 4D flow MRI

Four-dimensional flow magnetic resonance imaging (MRI) has evolved as a non-invasive imaging technique to visualize and quantify blood flow in the heart and vessels. Hemodynamic parameters derived from 4D flow MRI, such as net flow and peak velocities, but also kinetic energy, turbulent kinetic energy, viscous energy loss, and wall shear stress have shown to be of diagnostic relevance for cardiovascular diseases. 4D flow MRI, however, has several limitations. Its long acquisition times and its limited spatio-temporal resolutions lead to inaccuracies in velocity measurements in small and low-flow vessels and near the vessel wall. Additionally, 4D flow MRI requires long post-processing times, since inaccuracies due to the measurement process need to be corrected for and parameter quantification requires 2D and 3D contour drawing. Several machine learning (ML) techniques have been proposed to overcome these limitations. Existing scan acceleration methods have been extended using ML for image reconstruction and ML based super-resolution methods have been used to assimilate high-resolution computational fluid dynamic simulations and 4D flow MRI, which leads to more realistic velocity results. ML efforts have also focused on the automation of other post-processing steps, by learning phase corrections and anti-aliasing. To automate contour drawing and 3D segmentation, networks such as the U-Net have been widely applied. This review summarizes the latest ML advances in 4D flow MRI with a focus on technical aspects and applications. It is divided into the current status of fast and accurate 4D flow MRI data generation, ML based post-processing tools for phase correction and vessel delineation and the statistical evaluation of blood flow.

Clinical implications of the biomechanics of bicuspid aortic valve and bicuspid aortopathy

Bicuspid aortic valve (BAV), which affects up to 2% of the general population, results from the abnormal fusion of the cusps of the aortic valve. Patients with BAV are at a higher risk for developing aortic dilatation, a condition known as bicuspid aortopathy, which is associated with potentially life-threatening sequelae such as aortic dissection and aortic rupture. Although BAV biomechanics have been shown to contribute to aortopathy, their precise impact is yet to be delineated. Herein, we present the latest literature related to BAV biomechanics. We present the most recent definitions and classifications for BAV. We also summarize the current evidence pertaining to the mechanisms that drive bicuspid aortopathy. We highlight how aberrant flow patterns can contribute to the development of aortic dilatation. Finally, we discuss the role cardiac magnetic resonance imaging can have in assessing and managing patient with BAV and bicuspid aortopathy.