4D Flow Data Research Articles

Cardiothoracic Magnetic Resonance Flow Imaging

Multidimensional blood flow imaging with magnetic resonance has rapidly evolved over the last decade. The technique, often referred to as 4-dimensional (4D) flow, can now reliably image the heart and principal vessels of the chest in ≤15 minutes. In addition to dynamic 3D flow visualization, a range of unique quantitative hemodynamic markers can be calculated from 4D flow data. In this review article, we describe some of the more promising of these hemodynamic markers, including pulse wave velocity, pressure, turbulent kinetic energy, wall shear stress, and flow eccentricity. Evaluation of a range of cardiothoracic disorders has been explored with 4D flow, and many applications have been proposed. We also review the potential clinical applications of 4D flow in 4 broad contexts: the aorta, the pulmonary artery, acquired heart disease, and complex congenital heart disease. Promising preliminary results will be highlighted, including the use of abnormal systolic blood flow to risk-stratify patients for progressive valve-related aortic disease, turbulent kinetic energy to directly assess the hemodynamic impact of a stenotic lesion, and altered intracardiac flow to identify early heart failure. We discuss ongoing research efforts in the context of the larger clinical goals of 4D flow: the use of unique hemodynamic markers to (1) identify cardiovascular disease processes early in their course before clinical manifestation so that preemptive treatment can be undertaken; (2) refine the assessment of cardiovascular disease so as to better identify optimal medical or surgical therapies; and (3) enhance the evaluation and monitoring of the hemodynamic impact of different treatment options.

Improved visualization and quantification of 4D flow data using divergence-free wavelets

Background 4D flow MRI has the potential to provide global quantification of cardiac flow in a single acquisition [Hsiao, 2012]. However, 4D flow data are often compromised by low velocity-to-noise ratio, potentially caused by MRI acceleration or unfitting vencs. Since blood flow is approximately divergence-free, noise level can be reduced by removing divergence from noisy flow data [Song, 1993] [Busch, 2012]. On the other hand, strict enforcement of divergence-free condition distorts flow around edges as discrete approximation of flow near edges creates divergence. In this current work, we aim to provide an adjustable and fast operation of imposing multi-scale divergence-free conditions on flow data by using divergence-free wavelets. In addition, we utilize the sparsity of flow data in divergencefree wavelet domain [Deriaz, 2006] for further denoising by performing wavelet shrinkage [Donoho, 1995].

Fast semi-automated analysis of pulse wave velocity in the thoracic aorta using high temporal resolution 4D flow MRI

Background Pulse wave velocity (PWV) gives an indication of vessel stiffness, which can be used as a marker of age related changes in compliance, as well as to assess changes in vessel elasticity as a measure of atherosclerosis [1,2]. A recent meta-analysis showed PWV to be a robust predictor for cardiovascular events and all-cause mortality [3]. Recently, 4D flow MRI has been applied to assess PWV with full volumetric coverage of the aorta, but the analysis was limited by its low temporal resolution and the labor intensive segmentation of multiple analysis planes along the aorta. In this study, we propose a fast and semi-automated method for reliably extracting PWV from 4D flow data.

Reproducibility of quantitative analysis of aortic 4D flow data

Background 3D cine phase-contrast CMR (4D Flow”) permits quantitative assessment of anomalous alterations of aortic blood flow. Two hemodynamic parameters that have been used for this purpose is the wall shear stress (WSS), which is known to regulate endothelial cell function, and the normalized flow displacement from the vessel center, which was recently shown to correlate with increased growth rates of ascending aortic dilation [1,2]. Analysis of these hemodynamic parameters requires that a user 1) positions a 2D plane of interest in the volumetric 4D Flow dataset and 2) delineates the contour of the vascular lumen in this 2D plane. We set out to assess the reproducibility of 4D Flow-based estimation of WSS and normalized flow displacement at these two critical levels of user-interaction. Furthermore, we assessed which of the parameters correlate best with aortic growth.

Objective method for assessment of reliability of particle tracing visualization in 4D FLOW MRI

Background Three-dimensional (3D) velocity-encoded MRI (4DFLOW) of the heart may be used for the assessment of systolic and diastolic function. Particle tracing is frequently used as a technique for visualization of 3D intra-cardiac flow patterns. As the obtained velocity data may suffer from imperfections in the acquisition technique, it is important to have a means for image quality control. Consistency of the 4D FLOW data with other CMR measurements is a means for internal data consistency and may serve as surrogate validation. The purpose of the study was to apply particle tracing in 4DFLOW MRI of the heart and to compare aortic flow measurements derived from particle tracing with an alternative validated MR analysis technique. Methods

Comparison of 4D flow and 2D PC MRI blood flow quantification in children and young adults with congenital heart disease

Background Echocardiography (echo) is the primary imaging modality for assessment of aortic and pulmonary blood flow velocities. 2D phase contrast (PC) MRI provides better access to all segments of the aortic and pulmonary system and is considered the standard for evaluating blood flow. Both techniques are limited by velocity analysis in 2D planes and by single-direction velocity measurement which may be inadequate to characterize the complex 3D hemodynamics in congenital heart disease (CHD). 4D flow MRI provides simultaneous assessment of 3D blood flow characteristics of all vessels within a 3D volume and offers the ability to retrospectively quantify blood flow parameters at selectable regions of interest. The aim of this study is to test the potential of 4D flow for accuracy of quantification of aortic and pulmonary flow parameters compared to the reference standards echo and 2D PC MRI in children and young adults with CHD. Methods 32 patients with CHD who underwent simultaneous 4D flow and 2D PC MRI and echo within 9 months of MRI were retrospectively included. 2D PC MRI flow quantification in the aortic root (Ao), pulmonary trunk (PT), and right and left pulmonary arteries (RPA, LPA) was analyzed using Medis (Medis, Leiden, The Netherlands). 4D flow data analysis included calculation of a 3D-PC-angiogram which was used to position analysis planes in the Ao, PT, LPA and RPA (EnSight, CEI, Apex, NC) for quantification of net flow, regurgitant fraction, Qp:Qs, and peak velocities. Ao peak velocities were assessed by echo. Linear regression analysis was performed. Pearson’ sc orrelation coefficient (r) was calculated. A correlation with p<0.05 was considered significant. Results

Bicuspid Aortic Valve Is Associated With Altered Wall Shear Stress in the Ascending Aorta

Hemodynamics may play a role contributing to the progression of bicuspid aortic valve (BAV) aortopathy. This study measured the impact of BAV on the distribution of regional aortic wall shear stress (WSS) compared with control cohorts. Local WSS distribution was measured in the thoracic aorta of 60 subjects using 4-dimensional (4D) flow-sensitive magnetic resonance imaging. WSS analysis included 15 BAV patients: 12 with fusion of the right-left coronary cusp (6 stenotic) and 3 with fusion of the right and noncoronary cusp. The right-left BAV cohort was compared with healthy subjects (n=15), age-appropriate subjects (n=15), and age-/aorta size-controlled subjects (n=15). Compared with the age-appropriate and age-/aorta size-matched controls, WSS patterns in the right-left BAV ascending aorta were significantly elevated, independent of stenosis severity (peak WSS=0.9 ± 0.3 N/m(2) compared with 0.4 ± 0.3 N/m(2) in age-/aorta size-controlled subjects; P<0.001). Time-resolved (cine) 2D images of the bicuspid valves were coregistered with 4D flow data, directly linking cusp fusion pattern to a distinct ascending aortic flow jet pattern. The observation of right-anterior ascending aorta wall/jet impingement in right-left BAV patients corresponded to regions with statistically elevated WSS. Alternative jetting patterns were observed in the right and noncoronary cusp fusion patients. The results of this study demonstrate that bicuspid valves induced significantly altered ascending aorta hemodynamics compared with age- and size-matched controls with tricuspid valves. Specifically, the expression of increased and asymmetric WSS at the aorta wall was related to ascending aortic flow jet patterns, which were influenced by the BAV fusion pattern.

A finite-element approach to the direct computation of relative cardiovascular pressure from time-resolved MR velocity data

The evaluation of cardiovascular velocities, their changes through the cardiac cycle and the consequent pressure gradients has the capacity to improve understanding of subject-specific blood flow in relation to adjacent soft tissue movements. Magnetic resonance time-resolved 3D phase contrast velocity acquisitions (4D flow) represent an emerging technology capable of measuring the cyclic changes of large scale, multi-directional, subject-specific blood flow. A subsequent evaluation of pressure differences in enclosed vascular compartments is a further step which is currently not directly available from such data. The focus of this work is to address this deficiency through the development of a novel simulation workflow for the direct computation of relative cardiovascular pressure fields. Input information is provided by enhanced 4D flow data and derived MR domain masking. The underlying methodology shows numerical advantages in terms of robustness, global domain composition, the isolation of local fluid compartments and a treatment of boundary conditions. This approach is demonstrated across a range of validation examples which are compared with analytic solutions. Four subject-specific test cases are subsequently run, showing good agreement with previously published calculations of intra-vascular pressure differences. The computational engine presented in this work contributes to non-invasive access to relative pressure fields, incorporates the effects of both blood flow acceleration and viscous dissipation, and enables enhanced evaluation of cardiovascular blood flow.

Normal values of wall shear stress in the pulmonary artery from 4D flow data

m 2 ; RPA = 0.36±0.1N/m 2 ; LPA = 0.28±0.14N/m 2 . Figure 1 shows the average WSS-M calculated along the cardiac cycle for each segment. Figure 2 depicts Bland Altman plots of the mean WSS measured by the two observers, showing a small bias and standard deviations (mean difference of 0.016N/m 2 , 0.008N/m 2 and 0.005N/ m 2 for the WSS-M in the PA, RPA and LPA respectively). Conclusions In this work we proposed a novel and reproducible method to calculate WSS derived from 4D flow data in the main PA, RPA and LPA. In volunteers, we found a greater WSS in the RPA compared with the LPA, which is probably associated with more complex flow patterns (helices) in the RPA (2). Values of WSS obtained in patients showed increasing values of WSS, probably owe to complex and retrograde flow patterns in the pulmonary circulation.

Comparison of divergence-free algorithms for 3D MRI with three-directional velocity encoding

Summary Two divergence-free algorithms (finite difference and radial basis functions) are evaluated for improving the flow visualization quality of PC-VIPR 4D flow data. Both methods improve streamline length, with the finite difference method performing slightly better. Background The velocity-to-noise ratio in 4D MR Flow acquisitions can suffer, particularly for low velocity segments, because the velocity encoding settings are adjusted to the highest velocities in the volume. Blood is essentially incompressible so vascular velocity fields should be divergence free. Therefore, any remaining divergence in an acquired 3D or 4D MR flow dataset can be assumed to originate from image noise, which can be reduced by imposing divergence free conditions. Several methods have been proposed to reconstruct a divergence free velocity vector field, yet to our knowledge no comparison studies exist to explore their properties in flow MRI data. This study compares two methods in their effectiveness at improving flow visualizations, the finite difference method (FDM)[1] and the radial basis function (RBF)[2] based approach, which differ in their handling of noise and boundary conditions. Methods MR flow data were acquired in a straight tube flow phantom and the internal carotid arteries of a volunteer on a clinical 3T scanner with a 3D radially undersampled acquistion (PC-VIPR)[3]. Time resolved PCVIPR images were acquired as follows: scan time: 6 min acquired isotropic spatial resolution: 0.68 mm, TR/TE = 8.3/2.9 ms, temporal resolution: 39.5 ms. The FDM algorithm is solved with a fast Fourier solver for the Poisson equation. The RBF algorithm includes a normalized convolution, and is solved using iterative least squares. The quality of the resulting flow fields was quantified by average streamline lengths and streamline counts that extended from the emitter plane to a distal portion in the flow phantom or vessel. Data fidelity was measured by comparing flow values pre and post processing in the emitter and analysis planes. Results

An Information-Theoretic Framework for Flow Visualization

The process of visualization can be seen as a visual communication channel where the input to the channel is the raw data, and the output is the result of a visualization algorithm. From this point of view, we can evaluate the effectiveness of visualization by measuring how much information in the original data is being communicated through the visual communication channel. In this paper, we present an information-theoretic framework for flow visualization with a special focus on streamline generation. In our framework, a vector field is modeled as a distribution of directions from which Shannon's entropy is used to measure the information content in the field. The effectiveness of the streamlines displayed in visualization can be measured by first constructing a new distribution of vectors derived from the existing streamlines, and then comparing this distribution with that of the original data set using the conditional entropy. The conditional entropy between these two distributions indicates how much information in the original data remains hidden after the selected streamlines are displayed. The quality of the visualization can be improved by progressively introducing new streamlines until the conditional entropy converges to a small value. We describe the key components of our framework with detailed analysis, and show that the framework can effectively visualize 2D and 3D flow data.

Quantification of left ventricular kinetic energy using 4D flow MRI

Methods Six healthy volunteers age 32 ± 11 (mean ± SD), 4 men, were imaged in a 1.5 and 3 T Philips scanner, i.e. two scans per subject in one day. A 4D PC TFE-sequence with retrospective ECG-triggering and respiratory navigator, 3 mm isotropic voxels and a temporal resolution of 25 ms interpolated to 40 phases was used to quantify intraventricular flow. Typical sequence parameters: 1.5 T SENSE factor 2, TE/TR/flip (3.7/6.3 ms/8°). 3 T, SENSE factor 2, TE/TR/flip (3.7/6.2 ms/8°). Cine images from a SSFP sequence (TE/TR/flip, 1.5/2.9 ms/60°) with spatial resolution 1.48 × 1.48 mm, slice thickness 8 mm and a temporal resolution of 30 heart phases, covering the LV in short axis view was used to delineate the endocardial contours. 4D-flow data was post-processed using an in-house developed software including linear background phase correction. PC images were reconstructed from the 4Ddata in the short axis view and the endocardial contours were imported and manually corrected when needed. The kinetic energy was calculated as the squared velocity times the mass divided by two for all voxels within the LV for each time point. LV energy values are presented as mean ± SEM and results from the 1.5 T and 3 T measurements were compared according to Bland-Altman analysis (mean ± SD). from 13th Annual SCMR Scientific Sessions Phoenix, AZ, USA. 21-24 January 2010

Evaluation of Bicuspid Aortic Valve and Aortic Coarctation With 4D Flow Magnetic Resonance Imaging

Time-resolved, 3D, phase-contrast magnetic resonance imaging (4D flow) is an effective means of evaluating dynamic multidirectional blood flow in the thoracic aorta.1 We have used the technique for characterization of abnormal flow features in a 14-year-old boy with aortic coarctation and bicuspid aortic valve (BAV) but without evidence of aortic stenosis or regurgitation. In addition to the expected flow disturbance in the region of the juxtaductal coarctation (Figure 1), we show an unusual flow feature in the ascending aorta that has not been previously reported in this clinical setting and that may be unique to BAV: 2 discrete nested helices of midsystolic blood flow in a nonaneurysmal aorta (Figure 2). Figure 1. Fourteen-year-old boy with BAV and aortic coarctation. A, Three-dimensional contrast-enhanced magnetic resonance angiography that demonstrates a focal juxtaductal coarctation and prominent internal mammary and …

Turbulence Characteristics in Gradual Channel Transition

A field study was conducted to determine the effects of a channel transition on turbulence characteristics. Detailed three-dimensional (3D) flow measurements were collected at a cross section that is located downstream of a gradual channel expansion. These measurements were obtained via an acoustic doppler velocimeter and include the 3D velocity field, the mean local velocities, the turbulent intensities, the frictional characteristics of the flow, the secondary velocity along the transverse plane, and the instantaneous shear stress components in the streamwise and transverse directions. Analysis of the 3D flow data indicates that the turbulent flow on the outer bank of the channel is anisotropic. Such anisotropy of turbulence, which is attributed to the gradual expansion in the channel and bed roughness, yields the development of a secondary flow of Prandtl’s second kind as reported in 1952. In particular, it was found that turbulent intensities in the vertical and transverse directions on the outer bank...