Investigation of hemodynamic flow patterns caused by aortic stenosis using a combined 4D Flow MRI-CFD framework

Investigation of hemodynamic flow patterns caused by aortic stenosis using a combined 4D Flow MRI-CFD framework | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Investigation of hemodynamic flow patterns caused by aortic stenosis using a combined 4D Flow MRI-CFD framework Tianai Wang, Christine Quast, Florian Bönner, Malte Kelm, Tobias Zeus, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4593892/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose Aortic stenosis (AS) leads to alterations of supra-valvular flow patterns. These patterns might lead to, inter alia, an increased damage of red blood cell (RBC) membranes. The aim of this work was to elucidate these patient-specific patterns between a healthy subject and a patient suffering from severe AS through a 4D Flow MRI-based CFD methodology. Material and methods Computational models of subject-specific aortic geometries were created using in-vivo medical imaging data. Temporally and spatially resolved boundary conditions derived from 4D Flow MRI were implemented. After validation of the in-silico results with in-vivo data, the numerical flow fields were investigated regarding their blood flow characteristics, i.e. shear stresses on RBCs and helicity. These insights were used to determine the potential RBC damage in AS. Results The accuracy of the 4D Flow MRI-based CFD model was proven with excellent agreement between in-vivo and in-silico velocity fields and R² = 0.9. A pathological high shear stress region in the bulk flow was present during late systole with an increase of 125% compared to the healthy flow. The physiological bihelical structure with predominantly right-handed helices vanished for the pathological state. Instead, a left-handed helix appeared, accompanied by an overall increase in turbulent kinetic energy in areas of accumulated left-handed helicity. These alterations could cause RBC damage. Conclusion Validated 4D Flow MRI-based CFD models of healthy and AS patients suggest that altered turbulent and helical structures in the bulk flow are the cause for increased, potentially damaging forces acting upon RBCs in AS. Biomedical Engineering Computational fluid dynamics 4D Flow MRI aortic valve stenosis hemodynamics patientspecific modeling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction The development of cardiovascular diseases is linked to altered hemodynamic environments of the volumetric blood flow and therefore, gaining a deeper insight into both pathological and physiological flow conditions is of undeniable importance [1, 2]. A pathology of high prevalence and clinical relevance is aortic valve stenosis (AS) and the American Heart Society predicted an increase of more than 100 % in patients suffering from this disease by 2050 [3]. A better understanding of AS pathophysiology can help to develop safer and more reliable therapies to tackle these upcoming challenges. On this front, aortic flow patterns during AS play a decisive role in the pathology [4]. The presence of valvular pathologies leads to disturbed blood flow in the aorta which in turn has the potential to damage red blood cell (RBC) membranes. However, it remains uncertain which flow properties in particular are responsible for the damaging impact. In order to capture in-vivo flow features non-invasively, 4D Flow Magnetic Resonance Imaging (4D Flow MRI) is commonly used. 4D Flow MRI, however, does not provide sufficient temporal and spatial resolution due to its spatial- and phase-averaging nature to reliably derive all velocity scales and gradient-based hemodynamic parameters, such as shear stress fields on circulating particles [5], which are essential quantities to investigate the cause of blood damage [2, 6]. In order to overcome this limitation, patient-specific Computational Fluid Dynamics (CFD) has been gaining popularity in the field of medical engineering research. Here, the advantages of high spatial and temporal resolution of CFD simulations are combined with subject-specific information about geometry and boundary conditions. Different approaches have been developed over time, e.g. the use of realistic geometries segmented from medical images. More recent research is focusing on the definition of accurate boundary conditions, as they influence the agreement between the computed flow fields and reality [2]. Some studies used generic curves and uniform or parabolic inlet waveforms [7–10], or similarly simplified conditions at the outlets, such as constant or waveform pressure [7–9, 11]. These modeling strategies, however, neglect the spatial in-plane distribution of velocity which is highly complex and particularly characteristic for certain pathologies, such as AS. This complexity of in-vivo flow has led other studies to neglect the modeling of turbulence or the non-Newtonian behavior of blood in order to simplify the numerical problem [9–13]. In order to increase the realistic nature and the subject-specificity of the boundary conditions, the use of flow measurement techniques, such as Doppler echocardiography, time-resolved 2D Flow MRI or 4D Flow MRI, are the most promising approaches. While Doppler echocardiography can only provide one-directional flow information in two dimensions and the current gold standard for volumetric flow analysis of 2D Flow MRI is limited by a two-dimensional imaging plane, 4D Flow MRI is known to be the only existing method for non-invasive, time-resolved, and three-dimensional measurement of blood velocity in-vivo [2, 7]. Clinical flow data, especially derived from 4D Flow MRI, have not been used extensively to define boundary conditions so far. Instead, most existing studies using measured flow information as boundary conditions relied on in-vitro phantom data as these eliminate other uncertainties caused by the variability of humans [12]. The number of studies that have attempted to model the high complexity of real pathological or turbulent vascular flows (e.g., in the aortic arch during aortic valve stenosis) using in-vivo measurements for both geometry and boundary conditions is therefore limited [14]. Further on, a comparison to physiological flow structures and the relationship between aortic hemodynamics and RBC stresses during AS has not been performed in the past. However, bridging this gap is indispensable to continuously deepen the clinical knowledge and understanding of in-vivo vascular hemodynamics and their interplay with cardiovascular diseases. Therefore, the goal of the present work is to develop a 4D Flow MRI-based CFD methodology to generate patient-specific models of healthy and pathological aortic flow, and to show its ability to accurately characterize blood flow structures and alterations caused by AS in the aortic arch. By comparing healthy and pathological flow structures, hemodynamic differences between these flow fields will aid the identification of those flow features with potentially damaging effect on circulating erythrocytes during AS. Materials And Methods The complex aortic flow fields of a healthy adult and an adult suffering from severe AS were determined using a new methodology of CFD analysis based on 4D Flow MRI data and are described in detail in the following. The derived hemodynamics were compared using qualitative and quantitative methods to identify the largest differences in the aortic flow structures between the physiological and pathological conditions measured in this study. Study Design 4D Flow MRI and multisliced computed tomography (CT) measurements of a 78 year-old, female patient diagnosed with severe AS, originally acquired at the University Hospital Düsseldorf (UKD) were anonymized and made available for retrospective analysis. In addition, a 4D Flow MRI scan of a healthy (H) volunteer (female, age = 25 years) was acquired at the University Hospital Aachen (UKA) as physiological reference. Approvals were granted for all acquisitions by the respective Institutional Review Boards. 4D Flow data were acquired on 1.5 T MRI systems (Achieva at UKD and Ingenia Ambition at UKA, Philips Healthcare, The Netherlands) using the torso and posterior coil. Relevant scan parameters are listed in Supplement I (S-I). Single source CT data were acquired using a SOMATOM Definition AS+ (Siemens Healthcare, Forchheim, Germany), with a temporal resolution of 150 ms and collimation of 128 x 0.6 mm for the AS patient. In-Vivo Data Processing The handling of in-vivo data followed general guidelines and recommendations presented in the cardiovascular 4D Flow MRI consensus statement by Dyverfeldt, Bissell et al. [ 2 ]. The phase contrast DICOM images were processed using in-house developed MATLAB R2022a (MathWorks, Natick, MA) scripts. The data were first sorted according to timeframe and slice number. Furthermore, the stored greyscale values were transformed to their flow values in cm/s by multiplication with the velocity encoding value (V enc ). The magnitude images were used to segment the aortic geometry (aortic arch and supraaortic vessels) during systolic peak ejection with the open-source software ITK Snap ( www.itksnap.org ) [ 15 ]. For the AS case, the segmentation was done manually on a layer by layer basis by tracking the outer contour of the aortic arch. The magnitude images of the healthy volunteer provided sufficient signal-to-noise ratio and grey value contrast to enable semi-automated segmentation of all aortic parts, including the supra-aortic vessels, by interpolation between adjacent slices. Finally, the magnitude-based binary segmentations were superposed with the volumetric velocity data for each timeframe from the 4D Flow MRI scans using an in-house MATLAB script. This yielded a segmented velocity volume for the patient-specific lumen of the thoracic aorta. Resulting flow fields were visualized in ParaView 5.10.1 (KitWare, Clifton Park, NY). In-Silico Data Processing For subsequent generation of the numerical model, the commercially available CFD software package ANSYS CFX 2021 R2 was used, including the solid modeling Computer-Aided Design (CAD) software ANSYS SpaceClaim 2021 R2 and the meshing tool ANSYS ICEM CFD 2021 R2, all developed by ANSYS, Inc. (Canonsburg, PA). The vascular geometry for the AS case was semi-automatically segmented from CT images of the same patient. For the healthy case, the segmentation obtained from the 4D Flow MRI scans was used. Subsequently, the segmentation surfaces were manually smoothed using ANSYS SpaceClaim and all inlet and outlet surfaces were cut perpendicularly. Finally, the patient-specific geometries were meshed with suitable grid settings ensuring mesh independency of the numerical results. Meshing was performed with ANSYS ICEM CFD using unstructured tetrahedral meshes and prism elements for the near wall regions. A mesh independence study (see Supplement II (S-II) for further information) yielded a number of 4.65 million elements and 12 prism layers with a refinement region in the ascending aorta. Subsequently, the meshes were imported to ANSYS CFX. Here, the numerical model was built using patient-specific boundary conditions derived from the in-vivo measurements. The Carreau-Yasuda model was implemented for modeling of the Non-Newtonian blood behavior, utilizing parameter values defined by Abraham et al. [ 16 ] and a constant density ρ of 1060 kg/m 3 . Here, η 0 is the viscosity when the shear rate tends to zero, η ∞ is the viscosity for high shear rates, λ is a time constant, a and n are dimensionless parameters. Turbulent flow has been observed in AS in numerous studies, e.g. in a turbulence analysis by Manchester et al. [14]. Therefore, the Reynolds-averaged SST model was chosen to describe the expected formation of turbulent eddies. The inlet of the numerical model was located shortly above the aortic valve and the 4-dimensional velocity profile serving as input data was extracted from the corresponding plane in the 4D Flow MRI velocity volume. The velocity values were interpolated linearly in both space and time to adjust to the high-resolution CFD grid and time using an in-house MATLAB script. A preliminary analysis showed that the simulation of one and three cardiac cycles yielded similar results, indicating cycle-independency. Thus, only one cardiac cycle was simulated to reduce computational costs, with a numerical time step size of 1/20 of the corresponding MRI timeframe. The supra-aortic vessels and the descending aorta were treated as openings defined by loss coefficients, which control the resulting mass flow in the respective branch and serve as a substitute of the downstream vasculature and its impact on the flow structure. These loss coefficients can only be determined iteratively. Thus, for each numerical time step, an inner feedback loop was implemented with 100 iterations, which compared the percentage mass flow distribution through one opening for each iteration step with a prescribed percentage flow. According to the deviation, the loss coefficient was adapted for the next iteration until the two values were synchronized. This process was repeated for each numerical time step. The idealized percentage flow distributions were calculated from values provided in Benim et al. [17]. Finally, the vessel wall was modeled as a rigid, no-slip wall. Convergence control was set for a root mean square value of 10e-5. Numerical evaluation Finally, several parameters of interest were extracted from the CFD models for the validation of the model as well as for the subsequent investigation of the aortic flow behavior at characteristic timeframes. To improve comprehensibility, the following nomenclature for these timeframes T i,j was used, where a specific physical quantity “i” (“velocity” (V), “WSS” or “TSS”) reaches its peak value for the healthy (H) or aortic stenosis (AS) case “j”. The following cross-sectional and sagittal planes and volumes served as region of interest (ROI) for investigation: ROI A, ROI B, ROI C, SAG, and the AAo-Region, as shown in Fig. 1. Validation of the Numerical Model The temporal and spatial velocity fields were subjects of investigation for validating the generated CFD model. Therefore, the area-averaged velocity values on ROI A-C were compared between the CFD and MRI data for all 20 MRI timeframes by means of linear regression analysis. Further, spatial velocity distribution within these ROIs was analyzed qualitatively for the systolic timeframe. Parameters of Interest In order to locate the region of highest shear stress on erythrocytes potentially causing damage for the AS case, acting stresses and turbulence parameters were processed in both their temporal and spatial properties. The following quantities were of interest: Total shear stresses (TSS), together with Reynolds shear stresses (RSS) and wall shear stresses (WSS). Turbulent kinetic energy (TKE), helicity and its derivative, local normalized helicity (LNH). All these markers have been shown to play an important role in vascular flow studies. Their individual derivation and implementation is explained in the Supplementary Material (S-III). Results Validation of the In-Silico Model For a quantitative evaluation of the temporal evolution of velocity predicted by the CFD simulation, the area-averaged velocity at ROI A, B and C are plotted over the cardiac cycle for MRI and CFD and both cases, respectively. The temporal evolution for ROI B and C are shown in Fig. 2 a, b, d and e, the corresponding evolution at ROI A can be found in the Supplementary Material (S-IV). The noise magnitude of the MRI measurement equals to approx. 5% of V enc value [ 12 ] and is represented by the shaded areas. Hence, the numerical results show very good agreement concerning the shape and magnitude of the area-averaged velocity. Additionally, a linear regression analysis for ROI A-C and SAG deliver an R 2 -value of 0.9 for both the physiological and the pathological flow, indicating excellent agreement and statistically significant correlation between 4D Flow MRI-based CFD results and the measured in-vivo 4D Flow MRI velocities. The corresponding regression plots are presented in Fig. 2 c and f. Further, the numerically calculated and measured velocity fields also show high agreement spatially. Examples of spatial flow patterns and depictions of momentary regions of higher and lower velocities on ROI A-C are appended in the Supplement Material V (S-V) for the pathological flow field during early systole. Here, the difference between the high resolution CFD grid and the corresponding low-resolution MRI grid also becomes obvious. Physiological and Pathological In-Silico Velocity Fields With excellent agreement between the in-vivo and in-silico results, the in-silico velocity fields are investigated in more detail for the two numerical models. The numerically derived systolic aortic stenotic flow displays an eccentric high velocity jet flow along the outer wall of the ascending aorta. The velocity field of the healthy individual, however, remains rotationally symmetrical with a more parabolic profile. When comparing the maximum and averaged velocities of cross-sectional planes in the ascending aorta over a cardiac cycle, the area-averaged systolic value is higher in the healthy case, whereas the peak velocity is significantly higher for the aortic stenosis case, due to the stenotic jet. In the descending aorta, both average and maximum velocity is higher in physiological flow. Thus, the highest velocities are located in the ascending aorta for the AS patient and in the descending aorta for the healthy subject. Shear Stresses Acting onto Erythrocytes First, the total shear stress (TSS) and Reynolds shear stress (RSS) for physiological and pathological flow are compared on ROI A, as shown in Fig. 4. The area-averaged, temporal evolution shows, that TSS for the AS case is significantly higher than for the H case, with a second peak occurring during deceleration and being the global stress peak. At this point, it becomes 125 % higher than the TSS value in the healthy case. Further, RSS as an indicator for turbulent flow remains negligible for the H case but is clearly contributing to the overall acting stresses for AS. Contour II in Fig. 4 shows the eccentric region of high TSS in the pathological bulk flow at T TSS,AS , whereas the physiological stress field I does not show any obvious characteristics at its peak. The wall shear stress (WSS) evolution averaged for the entire ascending aortic wall is shown in the bottom part of Fig. 4. During systole, the average physiological WSS is larger than the pathological values. They become similar during diastole. During the initial deceleration phase, however, aortic valve stenosis leads to temporarily larger WSS. The corresponding WSS contours show high and nearly constant WSS distribution over the entire aortic wall for the healthy subject (III), whereas the stenotic flow shows locally high WSS in the ascending aorta with most wall areas only experiencing lower stress values (IV). With the significant rise in bulk RSS, bulk flow behavior is investigated in detail. First, turbulent kinetic energy (TKE) as a quantity of turbulence is compared qualitatively between physiological and pathological flow on SAG. The analysis shows a significant alteration for AS flow with swirling structures in the ascending aorta, see Fig. 5. Physiological flow does not show any regions of noticeable increase of TKE. Second, helicity and its locally normalized value, namely Local Normalized Helicity (LNH) were analyzed in the ascending aorta, as shown in Fig. 6. For the healthy subject, the helicity values remain positive over the entire cardiac cycle with their peak occurring shortly after the systolic velocity peak. It decreases with increasing distance from the aortic valve (from ROI A to ROI B). In AS, however, the mean helicity turns from left-handed to right-handed characteristics during late systole, and the magnitude reaches its peak later than in the healthy case. Fig. 6 c and d present the LNH contours on ROI A at their peak during deceleration. The healthy flow displays two counterrotating helices (c), while one nearly central vortex can be identified for the AS flow (d). When comparing this LNH contour with the corresponding TKE contour, the region of highest TKE coincides with the region of left-handed helicity (LNH < 0). Discussion The present study aimed to identify the alterations in aortic blood flow patterns caused by aortic valve stenosis which may lead to erythrocyte membrane damage. This was achieved by developing a subject-specific and realistic numerical flow model using in-vivo 4D Flow MRI. Flow Alterations during Aortic Stenosis and Erythrocyte Damage Velocity Field Altered by Aortic Stenosis Differences in the cross-sectional velocity distribution are caused by the presence of a high-velocity jet due to the reduced orifice of the aortic valve in AS. When analyzing the sagittal distribution, the age-difference between the healthy volunteer (age = 25 years) and the AS patient (age = 78 years) need to be taken into account. According to a velocity magnitude distribution analysis on healthy subjects, carried out by Garcia et al. [18], the velocity peak and profiles are highly age-dependent. Here, systolic velocity distribution for female subjects aged 21-39 years display a high velocity region not only in the supra-valvular region but also in the entire descending aorta. For individuals > 60 years however, the velocity in the descending aorta is significantly reduced and only the ascending aorta displays velocities > 1 m/s. [18] This change in velocity magnitude distribution is also observable in the present results and it therefore remains unclear if aortic valve stenosis further causes this decrease in high velocities in the descending aorta or if this phenomenon can be mainly traced back to the age-dependent evolution. Additionally, it is to be expected that this age-dependency also affects other hemodynamic parameters. However, subject-variability of aortic bulk flow properties, i.e. volumetric shear stresses and helicity, have not yet been the object of in-depth investigations in current research. Alterations in Derived Flow Quantities by Aortic Stenosis In order to pinpoint the impact of aortic flow alterations in AS on shear stress on circulating cells (e.g. erythrocytes), shear stress distributions, turbulence kinetic energy and helicity were investigated in detail. The results indicate that shear stresses acting onto the aortic wall (WSS) do not have significant damaging effects on erythrocytes in AS, as they are significantly higher for healthy flow during systole and remain similar during diastole compared to the AS flow. The TSS evolution, however, displays a key novel finding: it develops a second peak for the pathological flow during late systole, which shows an increase by 125 % in AS compared to the corresponding maximum physiological values. Therefore, the increased shear stresses acting in the bulk flow, which are indicative of the pathological state during AS, are likely to impact and apply damaging forces on circulating RBCs. Nonetheless, only the spatial characteristic of forces acting onto the erythrocytes is investigated in this present work. Flow-induced damage of erythrocytes, however, is an integrated effect of both spatial and temporal factors and Lagrangian-based analyses are required, since the instantaneous streamlines and particle pathlines do not align in complex aortic flow. Next, an additional investigation of the bulk flow structures during the TSS peak was carried out for deeper understanding. Here, an increase in turbulence characteristics in the bulk flow of the ascending aorta is present in AS, whereas TKE was negligible for the healthy case. Further, significant differences in helicity were identified for the ascending aortic bulk hemodynamics, which is a known parameter to evaluate blood flow alterations, particularly caused by valvular diseases [19, 20]. However, only little is known about the helical properties during AS [21]. According to Kilner et al. [22] and Markl et al. [23], two consistent features present in physiological flow of the human aorta are: helicity and retrograde flow. The anatomical curvature of the aortic arch leads to right-handed, well-aligned helical outflow during systolic peak. During deceleration, the ascending aorta is still predominantly filled with right-handed helices with a small portion of retrograde flow developing over a short distance during end-systole, which Morbiducci et al. [20] describe as a bihelical pattern with two counterrotating, Dean-like vortices, which is also characteristic for fully developed flow in a bended pipe. This can also be detected in the healthy flow structure of this study. The numerical results of the pathological AS flow in this study, however, show a decrease in right-handed helicity and an increase of the region of left-handed vortices. These areas of accumulated left-handed helices coincide with the areas of high TKE, thus pointing towards a correlation between left-handed helices and high presence of RBC-damaging turbulence potentially impacting erythrocyte function. Itatani et al. [19] state, that the goal of aortic valve surgeries is the reduction of helical patterns and the restoration of well-aligned flow streams. Overall, the results allow for the formulation of two main findings that – to the authors’ best knowledge – have not been identified in existing literature so far: In our AS case, the presence of a second TSS peak in the bulk flow during late systole/early diastole was observed, which correlates with the onset of strong turbulent structures in AS flow. This is not present in the healthy flow patterns and exceeds the corresponding physiological TSS peak by 125 % in our AS case. AS further causes a decrease in physiological right-handed helical structures. This also leads to the destruction of the bihelical structure, which is found to be characteristic for physiological aortic flow. Instead, a main left-handed helix is present, coinciding with the peak region of TKE. These two findings are expected to cause damage on circulating cells and cellular dysfunction. However, our work only included a size of N = 2. Therefore, these findings need to the further validated with a larger, age-matched group of both healthy individuals and AS patients. 4D Flow MRI-based CFD Methodology In order to validate the developed framework and the resulting findings, the contours of the numerical velocities profiles were analyzed on a set of planar, cross-sectional ROIs along the aortic centerline. Their temporal and spatial agreement with the corresponding velocity profiles obtained in the MRI process were compared qualitatively and quantitatively. A good qualitative agreement of patterns and contours is found on the ROIs. The linear regression analysis used to assess the overall correlation between the measured MRI velocity and the predicted CFD results yields excellent results with R 2 = 0.9 for both the physiological and the pathological case. To the best of the authors’ knowledge, this is the first study to compare patient-specific flow patterns in healthy flow and aortic stenosis by combining spatially and temporally resolved 4D Flow MRI measurements and CFD modeling in order to identify the pathological impact on erythrocytes due to flow alterations during AS. It implements realistic, four-dimensional in-vivo boundary conditions and the numerical modeling non-Newtonian and turbulent blood behavior. However, the 4D Flow MRI acquisition and post-processing suffered from limitations characteristic for this technique, e.g. the averaging nature of the scans, systematic errors [2], or segmentation uncertainties. Further, simplifying the vessel wall with rigid properties can lead to overestimations of some quantities such as the wall shear stress. According to Torii et al. [24], however, this overestimation becomes negligible (< 5 %) when analyzing time-averaged values and was therefore accepted in this study to keep computational costs low. Large differences between healthy flow and the altered hemodynamics present during AS were identified for the two individuals: The occurrence of a second TSS peak in the bulk flow of the ascending aorta during late systole and the loss of the physiological, predominantly right-handed bihelical structure with an increase in left-handed helices coinciding with areas of high TKE. These results were obtained by the development of a 4D Flow MRI-based CFD framework of high fidelity, regardless of the complexity of the in-vivo flow. These patient-specific models promise patient-individual treatment planning in the future, e.g. for aortic valve replacement, and improved long-term outcomes of patients undergoing such procedures. Declarations Funding: This research project was supported by the Studienstiftung des Deutschen Volkes (PhD Fellowship, German Academic Scholarship Foundation) and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) CRC/TRR259, Grant No. 397484323. Competing interests: The authors have no competing interests to declare that are relevant to the content of this article. Ethics approval: All acquisitions in this study were conducted for research purposes according to the Declaration of Helsinki following local ethical approval by the respective Institutional Review Boards of the University Hospital Düsseldorf (HHU, No. 2018-86, No. 5761R, No. 4080, R5761R) and the University Hospital Aachen (EK 038/22), Germany. Availability of data and material: The raw data can be retrieved upon reasonable request from the authors. Authors’ contributions: All authors contributed to the study conception and design. Creation of the numerical model, executing the simulations and data analysis were performed by TW. Result interpretation was performed by TW and MN. CQ, FB, MK, TZ and TL were involved in the acquisition of the clinical data. US and MN supervised the project. TW wrote the manuscript, considering the input of all co-authors. All authors read and approved the final version of the manuscript. References Long Q, Xu XY, Ariff B, Thom SA, Hughes AD, Stanton AV (2000) Reconstruction of blood flow patterns in a human carotid bifurcation: A combined CFD and MRI study. J Magn Reson Imaging 11:299–311. 10.1002/(SICI)1522-2586(200003)11:33.0.CO;2-M Dyverfeldt P, Bissell M, Barker AJ, Bolger AF, Carlhäll C-J, Ebbers T et al (2015) 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson 17:72. 10.1186/s12968-015-0174-5 Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS et al (2022) Heart Disease and Stroke Statistics-2022 Update: A Report From the American Heart Association. Circulation 145:e153–e639. 10.1161/CIR.0000000000001052 Gülan U, Lüthi B, Holzner M, Liberzon A, Tsinober A, Kinzelbach W (2014) An in vitro investigation of the influence of stenosis severity on the flow in the ascending aorta. Med Eng Phys 36:1147–1155. 10.1016/j.medengphy.2014.06.018 Glor FP, Westenberg JJM, Vierendeels J, Danilouchkine M, Verdonck P (2002) Validation of the coupling of magnetic resonance imaging velocity measurements with computational fluid dynamics in a U bend. Artif Organs 26:622–635. 10.1046/j.1525-1594.2002.07085.x Faghih MM, Sharp MK (2019) Modeling and prediction of flow-induced hemolysis: a review. Biomech Model Mechanobiol 18:845–881. 10.1007/s10237-019-01137-1 Miyazaki S, Itatani K, Furusawa T, Nishino T, Sugiyama M, Takehara Y, Yasukochi S (2017) Validation of numerical simulation methods in aortic arch using 4D Flow MRI. Heart Vessels 32:1032–1044. 10.1007/s00380-017-0979-2 Numata S, Itatani K, Kanda K, Doi K, Yamazaki S, Morimoto K et al (2016) Blood flow analysis of the aortic arch using computational fluid dynamics. Eur J Cardiothorac Surg 49:1578–1585. 10.1093/ejcts/ezv459 Steinman DA, Thomas JB, Ladak HM, Milner JS, Rutt BK, Spence JD (2002) Reconstruction of carotid bifurcation hemodynamics and wall thickness using computational fluid dynamics and MRI. Magn Reson Med 47:149–159. 10.1002/mrm.10025 Casacuberta SOUDAHE, GAMEZ-MONTERO PJ J, PÉREZ JS, RODRÍGUEZ-CANCIO M, RAUSH G et al (2017) Estimation of wall shear stress using 4D Flow cardiovascular MRI and computational fluid dynamics. J Mech Med Biol 17:1750046. 10.1142/S0219519417500464 Ferdian E, Suinesiaputra A, Dubowitz DJ, Zhao D, Wang A, Cowan B, Young AA (2020) 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics. Front Phys. 10.3389/fphy.2020.00138 Papathanasopoulou P, Zhao S, Köhler U, Robertson MB, Long Q, Hoskins P et al (2003) MRI measurement of time-resolved wall shear stress vectors in a carotid bifurcation model, and comparison with CFD predictions. J Magn Reson Imaging 17:153–162. 10.1002/jmri.10243 Köhler U, Marshall I, Robertson MB, Long Q, Xu XY, Hoskins PR (2001) MRI measurement of wall shear stress vectors in bifurcation models and comparison with CFD predictions. J Magn Reson Imaging 14:563–573. 10.1002/jmri.1220 Manchester EL, Pirola S, Salmasi MY, O'Regan DP, Athanasiou T, Xu XY (2021) Analysis of Turbulence Effects in a Patient-Specific Aorta with Aortic Valve Stenosis. Cardiovasc Eng Technol 12:438–453. 10.1007/s13239-021-00536-9 Yushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, Gerig G (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. NeuroImage 31:1116–1128. 10.1016/j.neuroimage.2006.01.015 Abraham F, Behr M, Heinkenschloss M (2005) Shape optimization in steady blood flow: a numerical study of non-Newtonian effects. Comput Methods Biomech Biomed Engin 8:127–137. 10.1080/10255840500180799 Benim AC, Nahavandi A, Assmann A, Schubert D, Feindt P, Suh SH (2011) Simulation of blood flow in human aorta with emphasis on outlet boundary conditions. Appl Math Model 35:3175–3188. 10.1016/j.apm.2010.12.022 Garcia J, van der Palen RLF, Bollache E, Jarvis K, Rose MJ, Barker AJ et al (2018) Distribution of blood flow velocity in the normal aorta: Effect of age and gender. J Magn Reson Imaging 47:487–498. 10.1002/jmri.25773 Itatani K, Sekine T, Yamagishi M, Maeda Y, Higashitani N, Miyazaki S et al (2022) Hemodynamic Parameters for Cardiovascular System in 4D Flow MRI: Mathematical Definition and Clinical Applications. Magn Reson Med Sci 21:380–399. 10.2463/mrms.rev.2021-0097 Morbiducci U, Ponzini R, Rizzo G, Cadioli M, Esposito A, de Cobelli F et al (2009) In vivo quantification of helical blood flow in human aorta by time-resolved three-dimensional cine phase contrast magnetic resonance imaging. Ann Biomed Eng 37:516–531. 10.1007/s10439-008-9609-6 von Knobelsdorff-Brenkenhoff F, Karunaharamoorthy A, Trauzeddel RF, Barker AJ, Blaszczyk E, Markl M, Schulz-Menger J (2016) Aortic flow and wall shear stress in aortic stenosis is associated with left ventricular remodeling. J Cardiovasc Magn Reson. 10.1186/1532-429X-18-S1-Q57 Kilner PJ, Yang GZ, Mohiaddin RH, Firmin DN, Longmore DB (1993) Helical and retrograde secondary flow patterns in the aortic arch studied by three-directional magnetic resonance velocity mapping. Circulation 88:2235–2247. 10.1161/01.cir.88.5.2235 Markl M, Draney MT, Hope MD, Levin JM, Chan FP, Alley MT et al (2004) Time-resolved 3-dimensional velocity mapping in the thoracic aorta: visualization of 3-directional blood flow patterns in healthy volunteers and patients. J Comput Assist Tomogr 28:459–468. 10.1097/00004728-200407000-00005 Torii R, Wood NB, Hadjiloizou N, Dowsey AW, Wright AR, Hughes AD et al (2009) Fluid–structure interaction analysis of a patient-specific right coronary artery with physiological velocity and pressure waveforms. Commun Numer Meth Eng 25:565–580. 10.1002/cnm.1231 Additional Declarations The authors declare no competing interests. Supplementary Files SUPPLEMENTARYMATERIAL.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4593892","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":315387357,"identity":"c34b8579-38d4-417d-9afb-b84977bcacfa","order_by":0,"name":"Tianai Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYLCCD1CaGUxKgAg2/DoYZ0DpZqK1MPOQpMXg+NnDr21qtskxSJ8xf1zAYCfPP7v52AOGMhvcWs7kpVnnHLttzMCXY9g8gyHZcMadY+kGDOfScGoxO5BjZpzDdjuxgYd3YzPvP2bGDRI5ZhKMbYdxazn/xszY4t/terAWHoZ6+w0S+d+AWv7j1nIjx/gxY9vtBAaIlsOJQFvYgFoO4NRif+ONGWNv323DNh7+j7N5GI4nz7iRZiaRcC4ZpxbJ/hzjDz++3Zbn52FL+MzDUG3bPyP5mcSHMjucWoCADUtEJODTAIzJD/jlR8EoGAWjYMQDAHccT0GjlHHUAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0000-5851-3915","institution":"Department of Cardiovascular Engineering, Institute of Applied Medical Engineering, Medical Faculty, RWTH Aachen University, Aachen, Germany","correspondingAuthor":true,"prefix":"","firstName":"Tianai","middleName":"","lastName":"Wang","suffix":""},{"id":315391631,"identity":"b718cb4b-f2b5-4d67-aae8-c4d92eb68fe2","order_by":1,"name":"Christine Quast","email":"","orcid":"","institution":"Department of Cardiology, Pulmonary Diseases and Vascular Medicine, Heinrich-Heine University, Düsseldorf, Germany","correspondingAuthor":false,"prefix":"","firstName":"Christine","middleName":"","lastName":"Quast","suffix":""},{"id":315391632,"identity":"1fef188d-1bd2-42ab-8621-a2945a47148b","order_by":2,"name":"Florian Bönner","email":"","orcid":"","institution":"Department of Cardiology, Pulmonary Diseases and Vascular Medicine, Heinrich-Heine University, Düsseldorf, Germany","correspondingAuthor":false,"prefix":"","firstName":"Florian","middleName":"","lastName":"Bönner","suffix":""},{"id":315391633,"identity":"c851b8e8-2663-48ca-91f3-00a77a058f4e","order_by":3,"name":"Malte Kelm","email":"","orcid":"","institution":"Department of Cardiology, Pulmonary Diseases and Vascular Medicine, Heinrich-Heine University, Düsseldorf, Germany","correspondingAuthor":false,"prefix":"","firstName":"Malte","middleName":"","lastName":"Kelm","suffix":""},{"id":315391634,"identity":"41326c83-ae89-4d41-9138-65b774c020ab","order_by":4,"name":"Tobias Zeus","email":"","orcid":"","institution":"Department of Cardiology, Pulmonary Diseases and Vascular Medicine, Heinrich-Heine University, Düsseldorf, Germany","correspondingAuthor":false,"prefix":"","firstName":"Tobias","middleName":"","lastName":"Zeus","suffix":""},{"id":315391635,"identity":"242d9af4-6898-445b-93e7-240292feed78","order_by":5,"name":"Teresa Lemainque","email":"","orcid":"","institution":"Department of Diagnostic and Interventional Radiology, Medical Faculty, RWTH Aachen University, Aachen, Germany","correspondingAuthor":false,"prefix":"","firstName":"Teresa","middleName":"","lastName":"Lemainque","suffix":""},{"id":315391636,"identity":"7c9f30cc-861c-41f2-8a67-03443d804814","order_by":6,"name":"Ulrich Steinseifer","email":"","orcid":"","institution":"Department of Cardiovascular Engineering, Institute of Applied Medical Engineering, Medical Faculty, RWTH Aachen University, Aachen, Germany","correspondingAuthor":false,"prefix":"","firstName":"Ulrich","middleName":"","lastName":"Steinseifer","suffix":""},{"id":315391637,"identity":"051bbdb1-557a-471b-82ce-cbcdc7dab2f9","order_by":7,"name":"Michael Neidlin","email":"","orcid":"","institution":"Department of Cardiovascular Engineering, Institute of Applied Medical Engineering, Medical Faculty, RWTH Aachen University, Aachen, Germany","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Neidlin","suffix":""}],"badges":[],"createdAt":"2024-06-17 11:37:58","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4593892/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4593892/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":58673720,"identity":"d4120441-13dc-4255-9259-46b3760bf1da","added_by":"auto","created_at":"2024-06-19 15:20:59","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":114720,"visible":true,"origin":"","legend":"\u003cp\u003eLocation of ROIs in the aortic model which are used for validation and post-processing of the numerical results\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/7c11e9e9b4ccab1a60281f15.png"},{"id":58672898,"identity":"8e0a364d-c27c-49e5-b830-2adf7cff201e","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":98610,"visible":true,"origin":"","legend":"\u003cp\u003eValidation results of the 4D Flow MRI-based CFD model, each plot compares the 4D Flow MRI measured values with the corresponding numerically determined velocities. (a-b) Area-averaged temporal evolution of velocity for H case on ROI B and C, respectively. (c) Linear regression plot for H velocities on ROI A-C and SAG for area-averaged velocities throughout the cardiac cycle. (d-e) Area-averaged temporal evolution of velocity for AS case on ROI B and C, respectively. (f) Linear regression plot for AS velocities on ROI A-C and SAG for area-averaged velocities throughout the cardiac cycle. Abbreviations: H – healthy, AS – patient with aortic stenosis, ROI – region of interest\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/eab3df244a5eb5c8f705f77f.png"},{"id":58672900,"identity":"cc79162a-5bad-471f-8556-f6fcc57cef7c","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":543725,"visible":true,"origin":"","legend":"\u003cp\u003eVelocity contours on sagittal and cross sectional plane ROI A at systolic peak for a) healthy case and b) aortic valve stenosis case\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/3ac9ca26d8ee1e3d4ca0e46b.png"},{"id":58672901,"identity":"e05b36e2-8dc7-4239-a6f2-5fb1cfd2cf03","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":364846,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal shear stress evolutions and contours at their respective peak time frames. (I) TSS contour on ROI A for healthy case, (II) TSS contour on ROI A for pathological case, (III) WSS contour on aortic wall for healthy case, (IV) WSS contour on aortic wall for pathological case\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/ca522ef56c33c05f269df122.png"},{"id":58672899,"identity":"8fb09072-be47-4b9b-bbd0-3a73aab98849","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":192524,"visible":true,"origin":"","legend":"\u003cp\u003eTurbulent kinetic energy distribution on sagittal plane SAG during deceleration for (a) physiological and (b) pathological flow\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/d20e801ba2952a8f4820531a.png"},{"id":58672904,"identity":"921bad76-5940-418c-8b31-49be5a84577d","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":464080,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal and spatial helicity distribution. (a) Area-averaged helicity evolution on ROI A for H and AS. (b) Area-averaged helicity evolution on ROI B for H and AS. (c-d) LNH contour and tangentially projected streamline vectors on ROI A during deceleration peak for H and AS, respectively. (e) TKE contour for AS case during TKE peak\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/04d2a46dadd0b81c9a43d1e9.png"},{"id":58673757,"identity":"80c97908-86a5-4c66-8245-c2234a984eb7","added_by":"auto","created_at":"2024-06-19 15:21:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1988108,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/af86e4fa-5db8-4132-b778-893ea4e6be25.pdf"},{"id":58672903,"identity":"56e42171-0231-4eea-a58f-19ddec40eff7","added_by":"auto","created_at":"2024-06-19 15:12:59","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3053713,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPLEMENTARYMATERIAL.docx","url":"https://assets-eu.researchsquare.com/files/rs-4593892/v1/73adc5012d66d62d28e1fc6c.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eInvestigation of hemodynamic flow patterns caused by aortic stenosis using a combined 4D Flow MRI-CFD framework\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe development of cardiovascular diseases is linked to altered hemodynamic environments of the volumetric blood flow and therefore, gaining a deeper insight into both pathological and physiological flow conditions is of undeniable importance [1, 2]. A pathology of high prevalence and clinical relevance is aortic valve stenosis (AS) and the American Heart Society predicted an increase of more than 100 % in patients suffering from this disease by 2050 [3]. A better understanding of AS pathophysiology can help to develop safer and more reliable therapies to tackle these upcoming challenges. On this front, aortic flow patterns during AS play a decisive role in the pathology [4]. The presence of valvular pathologies leads to disturbed blood flow in the aorta which in turn has the potential to damage red blood cell (RBC) membranes. However, it remains uncertain which flow properties in particular are responsible for the damaging impact. In order to capture in-vivo flow features non-invasively, 4D Flow Magnetic Resonance Imaging (4D Flow MRI) is commonly used. 4D Flow MRI, however, does not provide sufficient temporal and spatial resolution due to its spatial- and phase-averaging nature to reliably derive all velocity scales and gradient-based hemodynamic parameters, such as shear stress fields on circulating particles [5], which are essential quantities to investigate the cause of blood damage [2, 6].\u003c/p\u003e\n\u003cp\u003eIn order to overcome this limitation, patient-specific Computational Fluid Dynamics (CFD) has been gaining popularity in the field of medical engineering research. Here, the advantages of high spatial and temporal resolution of CFD simulations are combined with subject-specific information about geometry and boundary conditions. Different approaches have been developed over time, e.g. the use of realistic geometries segmented from medical images. More recent research is focusing on the definition of accurate boundary conditions, as they influence the agreement between the computed flow fields and reality [2]. Some studies used generic curves and uniform or parabolic inlet waveforms [7\u0026ndash;10], or similarly simplified conditions at the outlets, such as constant or waveform pressure [7\u0026ndash;9, 11]. These modeling strategies, however, neglect the spatial in-plane distribution of velocity which is highly complex and particularly characteristic for certain pathologies, such as AS. This complexity of in-vivo flow has led other studies to neglect the modeling of turbulence or the non-Newtonian behavior of blood in order to simplify the numerical problem [9\u0026ndash;13]. In order to increase the realistic nature and the subject-specificity of the boundary conditions, the use of flow measurement techniques, such as Doppler echocardiography, time-resolved 2D Flow MRI or 4D Flow MRI, are the most promising approaches. While Doppler echocardiography can only provide one-directional flow information in two dimensions and the current gold standard for volumetric flow analysis of 2D Flow MRI is limited by a two-dimensional imaging plane, 4D Flow MRI is known to be the only existing method for non-invasive, time-resolved, and three-dimensional measurement of blood velocity in-vivo [2, 7]. Clinical flow data, especially derived from 4D Flow MRI, have not been used extensively to define boundary conditions so far. Instead, most existing studies using measured flow information as boundary conditions relied on in-vitro phantom data as these eliminate other uncertainties caused by the variability of humans [12]. The number of studies that have attempted to model the high complexity of real pathological or turbulent vascular flows (e.g., in the aortic arch during aortic valve stenosis) using in-vivo measurements for both geometry and boundary conditions is therefore limited [14]. Further on, a comparison to physiological flow structures and the relationship between aortic hemodynamics and RBC stresses during AS has not been performed in the past. However, bridging this gap is indispensable to continuously deepen the clinical knowledge and understanding of in-vivo vascular hemodynamics and their interplay with cardiovascular diseases.\u003c/p\u003e\n\u003cp\u003eTherefore, the goal of the present work is to develop a 4D Flow MRI-based CFD methodology to generate patient-specific models of healthy and pathological aortic flow, and to show its ability to accurately characterize blood flow structures and alterations caused by AS in the aortic arch. By comparing healthy and pathological flow structures, hemodynamic differences between these flow fields will aid the identification of those flow features with potentially damaging effect on circulating erythrocytes during AS.\u003c/p\u003e"},{"header":"Materials And Methods","content":"\u003cp\u003eThe complex aortic flow fields of a healthy adult and an adult suffering from severe AS were determined using a new methodology of CFD analysis based on 4D Flow MRI data and are described in detail in the following. The derived hemodynamics were compared using qualitative and quantitative methods to identify the largest differences in the aortic flow structures between the physiological and pathological conditions measured in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy Design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e4D Flow MRI and multisliced computed tomography (CT) measurements of a 78 year-old, female patient diagnosed with severe AS, originally acquired at the University Hospital D\u0026uuml;sseldorf (UKD) were anonymized and made available for retrospective analysis. In addition, a 4D Flow MRI scan of a healthy (H) volunteer (female, age\u0026thinsp;=\u0026thinsp;25 years) was acquired at the University Hospital Aachen (UKA) as physiological reference. Approvals were granted for all acquisitions by the respective Institutional Review Boards. 4D Flow data were acquired on 1.5 T MRI systems (Achieva at UKD and Ingenia Ambition at UKA, Philips Healthcare, The Netherlands) using the torso and posterior coil. Relevant scan parameters are listed in Supplement I (S-I). Single source CT data were acquired using a SOMATOM Definition AS+ (Siemens Healthcare, Forchheim, Germany), with a temporal resolution of 150 ms and collimation of 128 x 0.6 mm for the AS patient.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIn-Vivo Data Processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe handling of in-vivo data followed general guidelines and recommendations presented in the cardiovascular 4D Flow MRI consensus statement by Dyverfeldt, Bissell et al. [\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e].\u003c/p\u003e\n\u003cp\u003eThe phase contrast DICOM images were processed using in-house developed MATLAB R2022a (MathWorks, Natick, MA) scripts. The data were first sorted according to timeframe and slice number. Furthermore, the stored greyscale values were transformed to their flow values in cm/s by multiplication with the velocity encoding value (V\u003csub\u003eenc\u003c/sub\u003e).\u003c/p\u003e\n\u003cp\u003eThe magnitude images were used to segment the aortic geometry (aortic arch and supraaortic vessels) during systolic peak ejection with the open-source software ITK Snap (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.itksnap.org\u003c/span\u003e\u003c/span\u003e) [\u003cspan class=\"CitationRef\"\u003e15\u003c/span\u003e]. For the AS case, the segmentation was done manually on a layer by layer basis by tracking the outer contour of the aortic arch. The magnitude images of the healthy volunteer provided sufficient signal-to-noise ratio and grey value contrast to enable semi-automated segmentation of all aortic parts, including the supra-aortic vessels, by interpolation between adjacent slices.\u003c/p\u003e\n\u003cp\u003eFinally, the magnitude-based binary segmentations were superposed with the volumetric velocity data for each timeframe from the 4D Flow MRI scans using an in-house MATLAB script. This yielded a segmented velocity volume for the patient-specific lumen of the thoracic aorta. Resulting flow fields were visualized in ParaView 5.10.1 (KitWare, Clifton Park, NY).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIn-Silico Data Processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor subsequent generation of the numerical model, the commercially available CFD software package ANSYS CFX 2021 R2 was used, including the solid modeling Computer-Aided Design (CAD) software ANSYS SpaceClaim 2021 R2 and the meshing tool ANSYS ICEM CFD 2021 R2, all developed by ANSYS, Inc. (Canonsburg, PA).\u003c/p\u003e\n\u003cp\u003eThe vascular geometry for the AS case was semi-automatically segmented from CT images of the same patient. For the healthy case, the segmentation obtained from the 4D Flow MRI scans was used. Subsequently, the segmentation surfaces were manually smoothed using ANSYS SpaceClaim and all inlet and outlet surfaces were cut perpendicularly. Finally, the patient-specific geometries were meshed with suitable grid settings ensuring mesh independency of the numerical results. Meshing was performed with ANSYS ICEM CFD using unstructured tetrahedral meshes and prism elements for the near wall regions. A mesh independence study (see Supplement II (S-II) for further information) yielded a number of 4.65\u0026nbsp;million elements and 12 prism layers with a refinement region in the ascending aorta.\u003c/p\u003e\n\u003cp\u003eSubsequently, the meshes were imported to ANSYS CFX. Here, the numerical model was built using patient-specific boundary conditions derived from the in-vivo measurements.\u003c/p\u003e\n\u003cp\u003eThe Carreau-Yasuda model was implemented for modeling of the Non-Newtonian blood behavior, utilizing parameter values defined by Abraham et al. [\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e] and a constant density \u003cem\u003e\u0026rho;\u003c/em\u003e of 1060 kg/m\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAhYAAAAiCAIAAACMUQcWAAAKAklEQVR4Ae1d7bG0LAylBWuwBXugBGuwBTugAzuwAiuwARuwA3rgnXfPTJ5Mgqyr693r3vgLWYTk8HEgAdYlewwBQ8AQMAQMgUMIuENf2UeGgCFgCBgC34DAsizDMJQ1med5K41RSBk6+9UQMAQMga9FYBzHEAJXL8Y4DENd1zwypTRNU9d1IjKlZBSiMbEYQ8AQMAS+H4F5nr33XM8Y4zRNbds6l6GGEELf9zy9UYhAw14NAUPAEPgrCNR1PU2T1jaEkKWQGGNVVfM8808yVMN/trAhYAgYAobA9yEwjmOWJ1JKWxSSUmofD0fDKISjYWFDwBAwBP4EAk3TCCsWqV2gEPy0rislNgohKCxgCBgChsBfQcA517ZtVtsChWDtws1fRiFZDC3SEDAEDIGvRWCeZ+ec2ItF2hYoRH9oFEK4WcAQ+CQCMcYQwtbu+8OSxRjbtuXTRpFV/3hEpL1+NwKaCbi+RiEcDQsbAjdAIMbYNI3Y63JMblCRsFF0XTeOYzbDqqq4aTubxiK/DAGjkC+rUFPnryMQHs95FIZh8N7XdS08pTHGuq41VUzT1DTN+XIth9sh4JwTjYRUKKxCpmlyzvHpyCWGLJxY4cWQcCIwDEPbtm+ZfImc7/W6H7Hfr9cvr9N3QV1Wc1kW771zrqqqtm23jM6oTWy3p/F9WZamabA4CCFUj4d+LTcAdKUQgh4d+scjPm/blvfTl8QWWdnrvRDw3m/NHrquc85lh2WwC2+N76eQruuaplmWRQOK/iB+mue5aRp96FF//q0xBcRuqnKhTh17RBsdx7Gu6xjjdVpnoaZxUw+7JMlLTXdd16qquq6LMcJiQK4Ipv2/rge4qKy+72OM2DAzjiPCAitKnA1kKWSeZ3FrBaiLAC+InS3FIm+NwNa5EN5EdY9o21ZE/mvHb4EjhFAYBbL9MKWEpvx2R+JbNLo6kzJiV5d+Xf5bdZpdPmN0Q9u9TqQtqLuuWx7P1swLh62qqhKzn62m23UddbNlWZxzNExDO2Eo0CM+aANrF0EhWNzwfq5x0xmmlEBmHN5hGPitR0/F5t9a+AsQ2DqdXlDt2tPp67o652jCVZBD/wRK5OsjneYuMYWRSKhwBjGR1S98zdapppBxHDHv7vt+67jslnbvhfqYQ1uryY3F4zhqc8FTCoHFGcSD8BYC2fidFFLXNV/cPBU7W5ZF3hcBmEz3yz8MgzYXvXMV0nWdWCnvFy6lVFWVlu+lHH5J4v3j2knEfom+BTF0nWoKoc/FwErxhcB7ofbe8yG1UK74SajJparrWrdqoake8UMIRDx939OaRpS79aoz1KuQZVlEb30q9lZxFn9fBMZx5CvRgiLLsuiW/MI1i9M01XUNF9+6rrjKUezxgP1XCIGpZdM0tJbH1Fs0X9y+oiNFbidfY4zwFAEL3GlcOGJzrDjeFcs5aMR+RsItqcgrME1TjLHve/hyj60ss3X6KQrRUAsQsGeJU8jhpuucG4YBVdk0TQhhHEees6AQ7aXw3lN39d6HEObHI2TOvsYYsSmLehySiVK6rhNO/qdiZ4uzyLsjsCzLuLHhm1QrpNm1ClnXFb5B2Km89+u6ip3FsPmKRjk+HrEPDIO4NiujX4l2Tzq8JdD3PbgU3RLeFxiX35I/MtlJIVnEfkbCLWXbtsUBhRACdsrBpfHqLJjy13X6EQrJQk1CIgDCoIH+TNOF1t57NDbnHPEByhIUApC5FZfblCCYyEEIT6/CU8Irru97Pt/Ux0Geik2lWMAQIAR2UQhSox+St5x7/GiZTD2QCkCAZvpYgmQ952jBWzmIvqHdiby3iNLFa9u2VVUR26GLijRnXndSCDg4q+/VEpa1w8qDBCtcx1bOh678pKz+X/Ye2o2+VdB5qJEzZjnaVQiB0VQON10tvKCQlFLf99QgdfrzMYKlxnEUZw/PF2E5/E0EXqAQuA1pySN2KxYGRN4P+a4PgXiZQkTiM6/146EcCvujKU0hAMU1pfEYPoZSVgXEzkt4mHExUPLpqp5EkwpPA7pOz1DIFVDTriq4HHRN0eznjU1XUwgWf4cNhk8rQpxO995TR376rSUwBAoIvEAhMEBRXsMwOOdo9V0YEIlClmWpqmrLVKWHGyrrjQEMkXzG99RK/mrpJ6fGTyWEXwpjsbYHviqtSI+JAo2kWHoeHm50nZ6hECEq2hWJqn+lmHLjJM9B1p0OCnlv09UUAib7mTuycP6DwLGAIXAGgRcoRBg02rbl3u+yuRn90HtfmGehX20RzOFptUAHJgsad04OkSJzvO6kkC3EyhKCYODuFlWQFebVSGHWExOFV3PTdfoRCtmCOqUEBdEeChRypulq0LIUopNZjCHw+xHYSyHC85FSqutamFML03kctRXpBTpXDIiiCBid+eEDTLr5XB6LLTp0M45jVVUvGXN2Ugj2MXOTEaQVg7iQkG/3xOSaC6/1fTVGTBS6rquqijJZ1xVcDl8x4VlVVXalouv0IxSyBTWohTzVWxSi/6mNAEFAqykSiFejEAGIvd4Xgb0UgtGK1hBgFOEVL5xywGVBWysMwCc211+EqfB8YN8qlRVCwN4zjJXDMNCZZL0Jkr4Sgf0UkkWsLKF/PChRnFsWYhx7FWTZNA0nfu89Kh03TfV9j+3aWxfN6jotUAh2ilMb2yP/Gaixi5e2h6SU4CEQ+xevaLpGIXsq19LcAoG9FIJGT54PYWyBqrCxZGej2HJeQARzbcq/kPLkT845PvEXk27sV86qA1LZU/r+cS2L2FMJ+d6z/WXtkRwTBarB7NKT8hFVNo4j9zCllEQCfKgpBIdRsNTDHoT9d6btV19DjeUmZ6zs9tkrmq5RCLUiC9wdgb0UIvTc6gN93/NpHb5C7xU58FfsOBQDEE9wXZj226AI7z0tlTAC0ut+CnlJ2ixiPAct4XUUwsvNbtSmI9MpJVxZSJ8ICtmqU00hlMPVgadQawEuarpb3UcLYDGGwC9H4CCFcHOK0FBfhpo119BXuKeF7NEU/wMBvVEnhABJ5nnG2RFal+w3ZL0quUaMctASwnaEBPj1uqWbHunIqwwTX/944uPhF0wV6vSDFEK0t997dFHT1cBSjVvAELgXAgcpRFjMhc4wlMNEgNkot6fzxOU/XeAprwijJ9M6AxsrYZEn/zBuIC/re142jhjPTUsI2gC2+uJl/u35sPDKpJTADc65uq4xENMlKGT+KtfpZykESysiwjJE1zVdo5Ay8vbrjRA4QiHYx8KNyAWFMeQJx3sh/U/+1LYtt8z8ZNE7y8pKSJvEuNltZ4YvJdPO8Jc+zyZ+euIy+9VHIq9oulz9jyhlhRoC70XgCIW8VwLLzRAwBAwBQ+CmCBiF3LTiTGxDwBAwBD6PgFHI5+vAJDAEDAFD4KYIGIXctOJMbEPAEDAEPo/Af+/EBx28krdnAAAAAElFTkSuQmCC\" width=\"534\" height=\"34\"\u003e\u003c/p\u003e\n\u003cp\u003eHere, \u0026eta;\u003csub\u003e0\u003c/sub\u003e is the viscosity when the shear rate tends to zero, \u0026eta;\u003csub\u003e\u0026infin;\u003c/sub\u003e is the viscosity for high shear rates, \u0026lambda; is a time constant, \u003cem\u003ea\u003c/em\u003e and \u003cem\u003en\u003c/em\u003e are dimensionless parameters. Turbulent flow has been observed in AS in numerous studies, e.g. in a turbulence analysis by Manchester et al. [14]. Therefore, the Reynolds-averaged SST model was chosen to describe the expected formation of turbulent eddies.\u003c/p\u003e\n\u003cp\u003eThe inlet of the numerical model was located shortly above the aortic valve and the 4-dimensional velocity profile serving as input data was extracted from the corresponding plane in the 4D Flow MRI velocity volume. The velocity values were interpolated linearly in both space and time to adjust to the high-resolution CFD grid and time using an in-house MATLAB script. A preliminary analysis showed that the simulation of one and three cardiac cycles yielded similar results, indicating cycle-independency. Thus, only one cardiac cycle was simulated to reduce computational costs, with a numerical time step size of 1/20 of the corresponding MRI timeframe.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe supra-aortic vessels and the descending aorta were treated as openings defined by loss coefficients, which control the resulting mass flow in the respective branch and serve as a substitute of the downstream vasculature and its impact on the flow structure. These loss coefficients can only be determined iteratively. Thus, for each numerical time step, an inner feedback loop was implemented with 100 iterations, which compared the percentage mass flow distribution through one opening for each iteration step with a prescribed percentage flow. According to the deviation, the loss coefficient was adapted for the next iteration until the two values were synchronized. This process was repeated for each numerical time step. The idealized percentage flow distributions were calculated from values provided in Benim et al. [17].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFinally, the vessel wall was modeled as a rigid, no-slip wall. Convergence control was set for a root mean square value of 10e-5.\u003c/p\u003e\n\u003ch2\u003eNumerical evaluation\u003c/h2\u003e\n\u003cp\u003eFinally, several parameters of interest were extracted from the CFD models for the validation of the model as well as for the subsequent investigation of the aortic flow behavior at characteristic timeframes. To improve comprehensibility, the following nomenclature for these timeframes T\u003csub\u003ei,j\u003c/sub\u003e was used, where a specific physical quantity \u0026ldquo;i\u0026rdquo; (\u0026ldquo;velocity\u0026rdquo; (V), \u0026ldquo;WSS\u0026rdquo; or \u0026ldquo;TSS\u0026rdquo;) reaches its peak value for the healthy (H) or aortic stenosis (AS) case \u0026ldquo;j\u0026rdquo;. The following cross-sectional and sagittal planes and volumes served as region of interest (ROI) for investigation: ROI A, ROI B, ROI C, SAG, and the AAo-Region, as shown in Fig.\u0026nbsp;1.\u003c/p\u003e\n\u003ch3\u003eValidation of the Numerical Model\u003c/h3\u003e\n\u003cp\u003eThe temporal and spatial velocity fields were subjects of investigation for validating the generated CFD model. Therefore, the area-averaged velocity values on ROI A-C were compared between the CFD and MRI data for all 20 MRI timeframes by means of linear regression analysis. Further, spatial velocity distribution within these ROIs was analyzed qualitatively for the systolic timeframe.\u003c/p\u003e\n\u003ch3\u003eParameters of Interest\u003c/h3\u003e\n\u003cp\u003eIn order to locate the region of highest shear stress on erythrocytes potentially causing damage for the AS case, acting stresses and turbulence parameters were processed in both their temporal and spatial properties. The following quantities were of interest: Total shear stresses (TSS), together with Reynolds shear stresses (RSS) and wall shear stresses (WSS). Turbulent kinetic energy (TKE), helicity and its derivative, local normalized helicity (LNH). All these markers have been shown to play an important role in vascular flow studies. Their individual derivation and implementation is explained in the Supplementary Material (S-III).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eValidation of the In-Silico Model\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor a quantitative evaluation of the temporal evolution of velocity predicted by the CFD simulation, the area-averaged velocity at ROI A, B and C are plotted over the cardiac cycle for MRI and CFD and both cases, respectively. The temporal evolution for ROI B and C are shown in Fig.\u0026nbsp;2 a, b, d and e, the corresponding evolution at ROI A can be found in the Supplementary Material (S-IV). The noise magnitude of the MRI measurement equals to approx. 5% of V\u003csub\u003eenc\u003c/sub\u003e value [\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e] and is represented by the shaded areas. Hence, the numerical results show very good agreement concerning the shape and magnitude of the area-averaged velocity. Additionally, a linear regression analysis for ROI A-C and SAG deliver an R\u003csup\u003e2\u003c/sup\u003e-value of 0.9 for both the physiological and the pathological flow, indicating excellent agreement and statistically significant correlation between 4D Flow MRI-based CFD results and the measured in-vivo 4D Flow MRI velocities. The corresponding regression plots are presented in Fig.\u0026nbsp;2 c and f.\u003c/p\u003e\n\u003cp\u003eFurther, the numerically calculated and measured velocity fields also show high agreement spatially. Examples of spatial flow patterns and depictions of momentary regions of higher and lower velocities on ROI A-C are appended in the Supplement Material V (S-V) for the pathological flow field during early systole. Here, the difference between the high resolution CFD grid and the corresponding low-resolution MRI grid also becomes obvious.\u003c/p\u003e\n\u003ch3\u003ePhysiological and Pathological In-Silico Velocity Fields\u003c/h3\u003e\n\u003cp\u003eWith excellent agreement between the in-vivo and in-silico results, the in-silico velocity fields are investigated in more detail for the two numerical models. The numerically derived systolic aortic stenotic flow displays an eccentric high velocity jet flow along the outer wall of the ascending aorta. The velocity field of the healthy individual, however, remains rotationally symmetrical with a more parabolic profile. When comparing the maximum and averaged velocities of cross-sectional planes in the ascending aorta over a cardiac cycle, the area-averaged systolic value is higher in the healthy case, whereas the peak velocity is significantly higher for the aortic stenosis case, due to the stenotic jet. In the descending aorta, both average and maximum velocity is higher in physiological flow. Thus, the highest velocities are located in the ascending aorta for the AS patient and in the descending aorta for the healthy subject.\u003c/p\u003e\n\u003ch3\u003eShear Stresses Acting onto Erythrocytes\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eFirst, the total shear stress (TSS) and Reynolds shear stress (RSS) for physiological and pathological flow are compared on ROI A, as shown in Fig.\u0026nbsp;4. The area-averaged, temporal evolution shows, that TSS for the AS case is significantly higher than for the H case, with a second peak occurring during deceleration and being the global stress peak. At this point, it becomes 125 % higher than the TSS value in the healthy case. Further, RSS as an indicator for turbulent flow remains negligible for the H case but is clearly contributing to the overall acting stresses for AS. Contour II in Fig.\u0026nbsp;4 shows the eccentric region of high TSS in the pathological bulk flow at T\u003csub\u003eTSS,AS\u003c/sub\u003e, whereas the physiological stress field I does not show any obvious characteristics at its peak.\u003c/p\u003e\n\u003cp\u003eThe wall shear stress (WSS) evolution averaged for the entire ascending aortic wall is shown in the bottom part of Fig.\u0026nbsp;4. During systole, the average physiological WSS is larger than the pathological values. They become similar during diastole. During the initial deceleration phase, however, aortic valve stenosis leads to temporarily larger WSS. The corresponding WSS contours show high and nearly constant WSS distribution over the entire aortic wall for the healthy subject (III), whereas the stenotic flow shows locally high WSS in the ascending aorta with most wall areas only experiencing lower stress values (IV).\u003c/p\u003e\n\u003cp\u003eWith the significant rise in bulk RSS, bulk flow behavior is investigated in detail. First, turbulent kinetic energy (TKE) as a quantity of turbulence is compared qualitatively between physiological and pathological flow on SAG. The analysis shows a significant alteration for AS flow with swirling structures in the ascending aorta, see Fig. 5. Physiological flow does not show any regions of noticeable increase of TKE.\u003c/p\u003e\n\u003cp\u003eSecond, helicity and its locally normalized value, namely Local Normalized Helicity (LNH) were analyzed in the ascending aorta, as shown in Fig.\u0026nbsp;6. For the healthy subject, the helicity values remain positive over the entire cardiac cycle with their peak occurring shortly after the systolic velocity peak.\u003csub\u003e\u0026nbsp;\u003c/sub\u003eIt decreases with increasing distance from the aortic valve (from ROI A to ROI B). In AS, however, the mean helicity turns from left-handed to right-handed characteristics during late systole, and the magnitude reaches its peak later than in the healthy case. Fig. 6 c and d present the LNH contours on ROI A at their peak during deceleration. The healthy flow displays two counterrotating helices (c), while one nearly central vortex can be identified for the AS flow (d). When comparing this LNH contour with the corresponding TKE contour, the region of highest TKE coincides with the region of left-handed helicity (LNH \u0026lt; 0).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study aimed to identify the alterations in aortic blood flow patterns caused by aortic valve stenosis which may lead to erythrocyte membrane damage. This was achieved by developing a subject-specific and realistic numerical flow model using in-vivo 4D Flow MRI.\u003c/p\u003e\n\u003ch2\u003eFlow Alterations during Aortic Stenosis and Erythrocyte Damage\u003c/h2\u003e\n\u003ch3\u003eVelocity Field Altered by Aortic Stenosis\u003c/h3\u003e\n\u003cp\u003eDifferences in the cross-sectional velocity distribution are caused by the presence of a high-velocity jet due to the reduced orifice of the aortic valve in AS. When analyzing the sagittal distribution, the age-difference between the healthy volunteer (age = 25 years) and the AS patient (age = 78 years) need to be taken into account. According to a velocity magnitude distribution analysis on healthy subjects, carried out by Garcia et al. [18], the velocity peak and profiles are highly age-dependent. Here, systolic velocity distribution for female subjects aged 21-39 years display a high velocity region not only in the supra-valvular region but also in the entire descending aorta. For individuals \u0026gt; 60 years however, the velocity in the descending aorta is significantly reduced and only the ascending aorta displays velocities \u0026gt; 1 m/s. [18] This change in velocity magnitude distribution is also observable in the present results and it therefore remains unclear if aortic valve stenosis further causes this decrease in high velocities in the descending aorta or if this phenomenon can be mainly traced back to the age-dependent evolution. Additionally, it is to be expected that this age-dependency also affects other hemodynamic parameters. However, subject-variability of aortic bulk flow properties, i.e. volumetric shear stresses and helicity, have not yet been the object of in-depth investigations in current research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAlterations in Derived Flow Quantities by Aortic Stenosis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn order to pinpoint the impact of aortic flow alterations in AS on shear stress on circulating cells (e.g. erythrocytes), shear stress distributions, turbulence kinetic energy and helicity were investigated in detail. The results indicate that shear stresses acting onto the aortic wall (WSS) do not have significant damaging effects on erythrocytes in AS, as they are significantly higher for healthy flow during systole and remain similar during diastole compared to the AS flow. The TSS evolution, however, displays a key novel finding: it develops a second peak for the pathological flow during late systole, which shows an increase by 125 % in AS compared to the corresponding maximum physiological values. Therefore, the increased shear stresses acting in the bulk flow, which are indicative of the pathological state during AS, are likely to impact and apply damaging forces on circulating RBCs. Nonetheless, only the spatial characteristic of forces acting onto the erythrocytes is investigated in this present work. Flow-induced damage of erythrocytes, however, is an integrated effect of both spatial and temporal factors and Lagrangian-based analyses are required, since the instantaneous streamlines and particle pathlines do not align in complex aortic flow.\u003c/p\u003e\n\u003cp\u003eNext, an additional investigation of the bulk flow structures during the TSS peak was carried out for deeper understanding. Here, an increase in turbulence characteristics in the bulk flow of the ascending aorta is present in AS, whereas TKE was negligible for the healthy case.\u003c/p\u003e\n\u003cp\u003eFurther, significant differences in helicity were identified for the ascending aortic bulk hemodynamics, which is a known parameter to evaluate blood flow alterations, particularly caused by valvular diseases [19, 20]. However, only little is known about the helical properties during AS [21]. According to Kilner et al. [22] and Markl et al. [23], two consistent features present in physiological flow of the human aorta are: helicity and retrograde flow. The anatomical curvature of the aortic arch leads to right-handed, well-aligned helical outflow during systolic peak. During deceleration, the ascending aorta is still predominantly filled with right-handed helices with a small portion of retrograde flow developing over a short distance during end-systole, which Morbiducci et al. [20] describe as a bihelical pattern with two counterrotating, Dean-like vortices, which is also characteristic for fully developed flow in a bended pipe. This can also be detected in the healthy flow structure of this study.\u003c/p\u003e\n\u003cp\u003eThe numerical results of the pathological AS flow in this study, however, show a decrease in right-handed helicity and an increase of the region of left-handed vortices. These areas of accumulated left-handed helices coincide with the areas of high TKE, thus pointing towards a correlation between left-handed helices and high presence of RBC-damaging turbulence potentially impacting erythrocyte function. Itatani et al. [19] state, that the goal of aortic valve surgeries is the reduction of helical patterns and the restoration of well-aligned flow streams.\u003c/p\u003e\n\u003cp\u003eOverall, the results allow for the formulation of two main findings that \u0026ndash; to the authors\u0026rsquo; best knowledge \u0026ndash; have not been identified in existing literature so far:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eIn our AS case, the presence of a second TSS peak in the bulk flow during late systole/early diastole was observed, which correlates with the onset of strong turbulent structures in AS flow. This is not present in the healthy flow patterns and exceeds the corresponding physiological TSS peak by 125 % in our AS case.\u003c/li\u003e\n \u003cli\u003eAS further causes a decrease in physiological right-handed helical structures. This also leads to the destruction of the bihelical structure, which is found to be characteristic for physiological aortic flow. Instead, a main left-handed helix is present, coinciding with the peak region of TKE.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThese two findings are expected to cause damage on circulating cells and cellular dysfunction. However, our work only included a size of N = 2. Therefore, these findings need to the further validated with a larger, age-matched group of both healthy individuals and AS patients.\u003c/p\u003e\n\u003ch2\u003e4D Flow MRI-based CFD Methodology\u003c/h2\u003e\n\u003cp\u003eIn order to validate the developed framework and the resulting findings, the contours of the numerical velocities profiles were analyzed on a set of planar, cross-sectional ROIs along the aortic centerline. Their temporal and spatial agreement with the corresponding velocity profiles obtained in the MRI process were compared qualitatively and quantitatively. A good qualitative agreement of patterns and contours is found on the ROIs. The linear regression analysis used to assess the overall correlation between the measured MRI velocity and the predicted CFD results yields excellent results with R\u003csup\u003e2\u003c/sup\u003e = 0.9 for both the physiological and the pathological case.\u003c/p\u003e\n\u003cp\u003eTo the best of the authors\u0026rsquo; knowledge, this is the first study to compare patient-specific flow patterns in healthy flow and aortic stenosis by combining spatially and temporally resolved 4D Flow MRI measurements and CFD modeling in order to identify the pathological impact on erythrocytes due to flow alterations during AS. It implements realistic, four-dimensional in-vivo boundary conditions and the numerical modeling non-Newtonian and turbulent blood behavior. However, the 4D Flow MRI acquisition and post-processing suffered from limitations characteristic for this technique, e.g. the averaging nature of the scans, systematic errors [2], or segmentation uncertainties. Further, simplifying the vessel wall with rigid properties can lead to overestimations of some quantities such as the wall shear stress. According to Torii et al. [24], however, this overestimation becomes negligible (\u0026lt; 5 %) when analyzing time-averaged values and was therefore accepted in this study to keep computational costs low.\u003c/p\u003e\n\u003cp\u003eLarge differences between healthy flow and the altered hemodynamics present during AS were identified for the two individuals: \u0026nbsp;The occurrence of a second TSS peak in the bulk flow of the ascending aorta during late systole and the loss of the physiological, predominantly right-handed bihelical structure with an increase in left-handed helices coinciding with areas of high TKE. These results were obtained by the development of a 4D Flow MRI-based CFD framework of high fidelity, regardless of the complexity of the in-vivo flow. These patient-specific models promise patient-individual treatment planning in the future, e.g. for aortic valve replacement, and improved long-term outcomes of patients undergoing such procedures.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research project was supported by the Studienstiftung des Deutschen Volkes (PhD Fellowship, German Academic Scholarship Foundation) and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) CRC/TRR259, Grant No. 397484323.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no competing interests to declare that are relevant to the content of this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll acquisitions in this study were conducted for research purposes according to the Declaration of Helsinki following local ethical approval by the respective Institutional Review Boards of the University Hospital D\u0026uuml;sseldorf (HHU, No. 2018-86, No. 5761R, No. 4080, R5761R) and the University Hospital Aachen (EK 038/22), Germany.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe raw data can be retrieved upon reasonable request from the authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors contributed to the study conception and design. Creation of the numerical model, executing the simulations and data analysis were performed by TW. Result interpretation was performed by TW and MN. CQ, FB, MK, TZ and TL were involved in the acquisition of the clinical data. US and MN supervised the project. TW wrote the manuscript, considering the input of all co-authors. All authors read and approved the final version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLong Q, Xu XY, Ariff B, Thom SA, Hughes AD, Stanton AV (2000) Reconstruction of blood flow patterns in a human carotid bifurcation: A combined CFD and MRI study. J Magn Reson Imaging 11:299\u0026ndash;311. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/(SICI)1522-2586(200003)11:3\u0026lt;299::AID-JMRI9\u0026gt;3.0.CO;2-M\u003c/span\u003e\u003cspan address=\"10.1002/(SICI)1522-2586(200003)11:3%3C299::AID-JMRI9%3E3.0.CO;2-M\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDyverfeldt P, Bissell M, Barker AJ, Bolger AF, Carlh\u0026auml;ll C-J, Ebbers T et al (2015) 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson 17:72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12968-015-0174-5\u003c/span\u003e\u003cspan address=\"10.1186/s12968-015-0174-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS et al (2022) Heart Disease and Stroke Statistics-2022 Update: A Report From the American Heart Association. Circulation 145:e153\u0026ndash;e639. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/CIR.0000000000001052\u003c/span\u003e\u003cspan address=\"10.1161/CIR.0000000000001052\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eG\u0026uuml;lan U, L\u0026uuml;thi B, Holzner M, Liberzon A, Tsinober A, Kinzelbach W (2014) An in vitro investigation of the influence of stenosis severity on the flow in the ascending aorta. Med Eng Phys 36:1147\u0026ndash;1155. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.medengphy.2014.06.018\u003c/span\u003e\u003cspan address=\"10.1016/j.medengphy.2014.06.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlor FP, Westenberg JJM, Vierendeels J, Danilouchkine M, Verdonck P (2002) Validation of the coupling of magnetic resonance imaging velocity measurements with computational fluid dynamics in a U bend. Artif Organs 26:622\u0026ndash;635. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1046/j.1525-1594.2002.07085.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1525-1594.2002.07085.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaghih MM, Sharp MK (2019) Modeling and prediction of flow-induced hemolysis: a review. Biomech Model Mechanobiol 18:845\u0026ndash;881. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10237-019-01137-1\u003c/span\u003e\u003cspan address=\"10.1007/s10237-019-01137-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiyazaki S, Itatani K, Furusawa T, Nishino T, Sugiyama M, Takehara Y, Yasukochi S (2017) Validation of numerical simulation methods in aortic arch using 4D Flow MRI. Heart Vessels 32:1032\u0026ndash;1044. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00380-017-0979-2\u003c/span\u003e\u003cspan address=\"10.1007/s00380-017-0979-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNumata S, Itatani K, Kanda K, Doi K, Yamazaki S, Morimoto K et al (2016) Blood flow analysis of the aortic arch using computational fluid dynamics. Eur J Cardiothorac Surg 49:1578\u0026ndash;1585. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/ejcts/ezv459\u003c/span\u003e\u003cspan address=\"10.1093/ejcts/ezv459\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteinman DA, Thomas JB, Ladak HM, Milner JS, Rutt BK, Spence JD (2002) Reconstruction of carotid bifurcation hemodynamics and wall thickness using computational fluid dynamics and MRI. Magn Reson Med 47:149\u0026ndash;159. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/mrm.10025\u003c/span\u003e\u003cspan address=\"10.1002/mrm.10025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCasacuberta SOUDAHE, GAMEZ-MONTERO PJ J, P\u0026Eacute;REZ JS, RODR\u0026Iacute;GUEZ-CANCIO M, RAUSH G et al (2017) Estimation of wall shear stress using 4D Flow cardiovascular MRI and computational fluid dynamics. J Mech Med Biol 17:1750046. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1142/S0219519417500464\u003c/span\u003e\u003cspan address=\"10.1142/S0219519417500464\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFerdian E, Suinesiaputra A, Dubowitz DJ, Zhao D, Wang A, Cowan B, Young AA (2020) 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics. Front Phys. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fphy.2020.00138\u003c/span\u003e\u003cspan address=\"10.3389/fphy.2020.00138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePapathanasopoulou P, Zhao S, K\u0026ouml;hler U, Robertson MB, Long Q, Hoskins P et al (2003) MRI measurement of time-resolved wall shear stress vectors in a carotid bifurcation model, and comparison with CFD predictions. J Magn Reson Imaging 17:153\u0026ndash;162. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/jmri.10243\u003c/span\u003e\u003cspan address=\"10.1002/jmri.10243\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eK\u0026ouml;hler U, Marshall I, Robertson MB, Long Q, Xu XY, Hoskins PR (2001) MRI measurement of wall shear stress vectors in bifurcation models and comparison with CFD predictions. J Magn Reson Imaging 14:563\u0026ndash;573. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/jmri.1220\u003c/span\u003e\u003cspan address=\"10.1002/jmri.1220\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManchester EL, Pirola S, Salmasi MY, O'Regan DP, Athanasiou T, Xu XY (2021) Analysis of Turbulence Effects in a Patient-Specific Aorta with Aortic Valve Stenosis. Cardiovasc Eng Technol 12:438\u0026ndash;453. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s13239-021-00536-9\u003c/span\u003e\u003cspan address=\"10.1007/s13239-021-00536-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, Gerig G (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. NeuroImage 31:1116\u0026ndash;1128. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuroimage.2006.01.015\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2006.01.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbraham F, Behr M, Heinkenschloss M (2005) Shape optimization in steady blood flow: a numerical study of non-Newtonian effects. Comput Methods Biomech Biomed Engin 8:127\u0026ndash;137. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/10255840500180799\u003c/span\u003e\u003cspan address=\"10.1080/10255840500180799\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenim AC, Nahavandi A, Assmann A, Schubert D, Feindt P, Suh SH (2011) Simulation of blood flow in human aorta with emphasis on outlet boundary conditions. Appl Math Model 35:3175\u0026ndash;3188. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.apm.2010.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.apm.2010.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia J, van der Palen RLF, Bollache E, Jarvis K, Rose MJ, Barker AJ et al (2018) Distribution of blood flow velocity in the normal aorta: Effect of age and gender. J Magn Reson Imaging 47:487\u0026ndash;498. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/jmri.25773\u003c/span\u003e\u003cspan address=\"10.1002/jmri.25773\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eItatani K, Sekine T, Yamagishi M, Maeda Y, Higashitani N, Miyazaki S et al (2022) Hemodynamic Parameters for Cardiovascular System in 4D Flow MRI: Mathematical Definition and Clinical Applications. Magn Reson Med Sci 21:380\u0026ndash;399. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2463/mrms.rev.2021-0097\u003c/span\u003e\u003cspan address=\"10.2463/mrms.rev.2021-0097\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorbiducci U, Ponzini R, Rizzo G, Cadioli M, Esposito A, de Cobelli F et al (2009) In vivo quantification of helical blood flow in human aorta by time-resolved three-dimensional cine phase contrast magnetic resonance imaging. Ann Biomed Eng 37:516\u0026ndash;531. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10439-008-9609-6\u003c/span\u003e\u003cspan address=\"10.1007/s10439-008-9609-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Knobelsdorff-Brenkenhoff F, Karunaharamoorthy A, Trauzeddel RF, Barker AJ, Blaszczyk E, Markl M, Schulz-Menger J (2016) Aortic flow and wall shear stress in aortic stenosis is associated with left ventricular remodeling. J Cardiovasc Magn Reson. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/1532-429X-18-S1-Q57\u003c/span\u003e\u003cspan address=\"10.1186/1532-429X-18-S1-Q57\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKilner PJ, Yang GZ, Mohiaddin RH, Firmin DN, Longmore DB (1993) Helical and retrograde secondary flow patterns in the aortic arch studied by three-directional magnetic resonance velocity mapping. Circulation 88:2235\u0026ndash;2247. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.cir.88.5.2235\u003c/span\u003e\u003cspan address=\"10.1161/01.cir.88.5.2235\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarkl M, Draney MT, Hope MD, Levin JM, Chan FP, Alley MT et al (2004) Time-resolved 3-dimensional velocity mapping in the thoracic aorta: visualization of 3-directional blood flow patterns in healthy volunteers and patients. J Comput Assist Tomogr 28:459\u0026ndash;468. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/00004728-200407000-00005\u003c/span\u003e\u003cspan address=\"10.1097/00004728-200407000-00005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTorii R, Wood NB, Hadjiloizou N, Dowsey AW, Wright AR, Hughes AD et al (2009) Fluid\u0026ndash;structure interaction analysis of a patient-specific right coronary artery with physiological velocity and pressure waveforms. Commun Numer Meth Eng 25:565\u0026ndash;580. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/cnm.1231\u003c/span\u003e\u003cspan address=\"10.1002/cnm.1231\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"1511c606-b0fe-41d8-a79e-1d93497e56f0","identifier":"10.13039/501100001659","name":"Deutsche Forschungsgemeinschaft","awardNumber":"397484323","order_by":0},{"identity":"34361ec6-e727-4e89-adb3-d990fe6f6065","identifier":"10.13039/501100004350","name":"Studienstiftung des Deutschen Volkes","awardNumber":"none","order_by":1}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"RWTH Aachen University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Computational fluid dynamics, 4D Flow MRI, aortic valve stenosis, hemodynamics, patientspecific modeling","lastPublishedDoi":"10.21203/rs.3.rs-4593892/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4593892/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eAortic stenosis (AS) leads to alterations of supra-valvular flow patterns. These patterns might lead to, inter alia, an increased damage of red blood cell (RBC) membranes. The aim of this work was to elucidate these patient-specific patterns between a healthy subject and a patient suffering from severe AS through a 4D Flow MRI-based CFD methodology.\u003c/p\u003e\u003ch2\u003eMaterial and methods\u003c/h2\u003e \u003cp\u003eComputational models of subject-specific aortic geometries were created using in-vivo medical imaging data. Temporally and spatially resolved boundary conditions derived from 4D Flow MRI were implemented. After validation of the in-silico results with in-vivo data, the numerical flow fields were investigated regarding their blood flow characteristics, i.e. shear stresses on RBCs and helicity. These insights were used to determine the potential RBC damage in AS.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe accuracy of the 4D Flow MRI-based CFD model was proven with excellent agreement between in-vivo and in-silico velocity fields and R\u0026sup2; = 0.9. A pathological high shear stress region in the bulk flow was present during late systole with an increase of 125% compared to the healthy flow. The physiological bihelical structure with predominantly right-handed helices vanished for the pathological state. Instead, a left-handed helix appeared, accompanied by an overall increase in turbulent kinetic energy in areas of accumulated left-handed helicity. These alterations could cause RBC damage.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eValidated 4D Flow MRI-based CFD models of healthy and AS patients suggest that altered turbulent and helical structures in the bulk flow are the cause for increased, potentially damaging forces acting upon RBCs in AS.\u003c/p\u003e","manuscriptTitle":"Investigation of hemodynamic flow patterns caused by aortic stenosis using a combined 4D Flow MRI-CFD framework","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-19 15:12:54","doi":"10.21203/rs.3.rs-4593892/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7784ac97-5289-47b5-8b04-73665f30a7c0","owner":[],"postedDate":"June 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":33349337,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2024-06-19T15:12:54+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-19 15:12:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4593892","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4593892","identity":"rs-4593892","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}