Registration for the conference "Modeling blood flow in the cardiovascular system and artificial devices - recent developments and challenges", Prague, March 27-29, 2026
The deadline for the registration is March 12, 2026.
The deadline for the registration is March 12, 2026.
Abstract: This presentation focuses on the estimation of cardiovascular model parameters, such as boundary conditions and mechanical properties, primarily from velocity-encoded MRI measurements. A significant portion of the talk addresses the consideration of imaging artifacts and their impact on parameter accuracy.
Abstract: This study uses computational fluid dynamics to investigate how blood rheology and wall boundary conditions affect hemodynamics in a cerebral aneurysm. Newtonian and Carreau models with hematocrit levels of 25%, 45%, and 65% were examined under no-slip and partial slip boundary conditions, and key metrics such as TAWSS, OSI, and LSA were evaluated. The results revealed substantial differences across the models considered, with wall slip having a stronger effect than the rheological model. These findings emphasize the need for careful selection of patient-specific modeling assumptions when using hemodynamic indicators for clinical interpretation.
Abstract: Patient-specific in-silico trials of the left atrium offer powerful predictive capabilities for stroke prevention and left atrial appendage occlusion (LAAO) optimization. This talk traces the evolution of these digital twins from large-scale hemodynamic robustness to advanced multi-therapeutic—device and drug—modeling. First, we will discuss the computational challenges and boundary condition strategies (specifically modeling atrial wall motion and inlet/outlet definitions) required to execute massive in-silico clinical trials, highlighting insights from the ~300-case IDEAL study. Building upon this foundation, the presentation will introduce a novel multiphysics framework that mathematically couples 3D computational fluid dynamics of device hemodynamics with 0D biochemical kinetic models of the coagulation cascade. By simultaneously simulating physical implants and pharmacological kinetics, this 0D-3D coupled approach provides a cutting-edge methodology to computationally tailor post-LAAO anticoagulant therapies while navigating the complex mathematical challenges of multiscale validation.
Abstract: Cardiovascular biomechanical modeling is an active area of research, with promising clinical proof-of-concept works. The models can be employed to augment the functional assessment of heart and vascular system – aiming for obtaining objective and reproducible characteristics –, or in planning the clinical management. These aspects will be exemplified on two groups of patients: (1) patients with tetralogy of Fallot (TOF) after complete surgical repair, suffering from chronic pulmonary valve insufficiency, who are scheduled for pulmonary valve replacement (PVR); and (2) patients born with a borderline-size ventricle, who are considered for a biventricular (BiV) repair (i.e., septation of the ventricle and separation of the systemic and pulmonary circulations), instead of the more common single-ventricle palliation (staged-surgery culminating in Fontan circulation). In the patients with TOF, the optimal timing of PVR will avoid irreversible remodeling of the ventricle. For the patients undergoing the BiV repair, the complex procedure brings the potential to completely avoid the long-term risks of the single-ventricle circulation (ventricular failure, pulmonary hypertension, liver fibrosis, etc.), however, with the immediate post-surgery risks of the borderline-size ventricle failure to sustain the whole circulation. As the heuristic criteria of direct measurements and their cut-off values address the question of the suitability for and associated risks of the given intervention only partially, within our interdisciplinary team – including pediatric cardiothoracic surgeons, cardiologists, radiologists, and biomechanical modelers – we use biomechanical models for synthesizing multi-modality data into a single framework; estimating biomarkers, that are not directly visible in the data; and simulating possible post-intervention scenarios. These three components of model prediction will be demonstrated during the presentation.
Abstract: The helicity of a divergence-free vector field is a standard measure for the extent to which the field lines wrap and coil around one another. Helicity is conserved in ideal incompressible systems; in the presence of viscosity, it instead follows an evolution law, with conservation recovered in the limit as the viscosity goes to zero. Standard discretisations of the incompressible Navier—Stokes do not discretely enforce this evolution law, and exhibit nonphysical helicity decay, artificially dissipating vortical structure.
In this talk we devise an arbitrary-order discretisation of the incompressible Navier—Stokes equations that does preserve the right helicity evolution law (as well as the right energy dissipation law). This discretisation gives substantially superior results to standard schemes on benchmark vortex simulations. Could this discretisation give better simulations of vortical structures in blood?
Abstract: This contribution provides an overview of over a decade of research focused on integrating the computationally efficient lattice Boltzmann method with experimental flow measurements obtained in flow phantoms via magnetic resonance imaging (MRI) for computational fluid dynamics (CFD) studies. Recently, this workflow was extended to include three-dimensional modeling of patient-specific vascular geometries and their fabrication using affordable 3D printing technologies. This talk will outline the challenges encountered during the development of this approach and summarize the main findings from each stage of the research. Particular attention will be given to the reliability of MRI-based flow measurements in turbulent regimes and the necessity of incorporating non-Newtonian blood rheology in flow simulations through the aortic valve. The presentation will also discuss ongoing efforts to design and produce custom-made, low-cost, MRI-compatible flow phantoms.
Abstract: In numerical simulations of blood flow in the native and mechanically assisted circulation, it is commonly assumed that blood adheres to the vessel walls, described by a well-known “no-slip” boundary condition. However, there are several reasons that suggest that assumption of the no-slip condition may be unfounded; these include: (1) variable wall properties, (2) the complex composition of blood as a heterogeneous mixture, and (3) the maximization of dissipation and entropy production under certain slip conditions. On the example of descending aorta, we demonstrate that using patient-specific MRI data, incorporating both vascular geometry (3D SSFP MRI) and measured velocity field (4D-PC MRI), we can determine the slip parameter at the boundary that best matches the observed velocity field.
Abstract: Low-gradient aortic stenosis represents (LGAS) a complex and diagnostically challenging subset of valvular heart disease, characterized by discordance between a small aortic valve area and a low transvalvular gradient. This clinical entity encompasses heterogeneous patient populations with both reduced and preserved left ventricular ejection fraction, and requires careful differentiation between true severe and pseudo-severe stenosis.
The diagnostic workup is complicated by limitations of conventional echocardiographic parameters, necessitating the integration of advanced imaging modalities and functional assessment. Therapeutic decision-making remains equally challenging, particularly in balancing the risks and benefits of aortic valve intervention in often frail and comorbid patients. This presentation will review the pathophysiological mechanisms underlying LGAS and highlight current diagnostic strategies.
Abstract: This lecture explores the aorta and aortic valve as critical components of the cardiovascular system, focusing on their anatomy, function, and the spectrum of genetic and congenital diseases that affect them. After a brief overview of normal physiology, the discussion will center on genetic aortic diseases—particularly those predisposing to life-threatening complications such as aortic dissection or rupture. The role of preventive surgical interventions on the thoracic aorta will be highlighted, emphasizing strategies to mitigate these risks. For the aortic valve, the lecture will address congenital abnormalities and their impact on valve function, as well as contemporary surgical approaches to correct valve dysfunction. A key theme throughout will be the essential role of magnetic resonance imaging in assessing blood flow, guiding diagnosis, and optimizing the timing of surgical intervention for both aortic and aortic valve diseases.
Abstract:
Abstract: The development of mechanical support systems for the human cardiovascular system, particularly Ventricular Assist Devices (VADs), presents significant engineering and biomedical challenges. To achieve clinical viability, these devices must combine high hydraulic performance and efficiency with compact size, low power consumption, and, crucially, long-term biocompatibility. Key risks such as hemolysis and thrombus formation must be mitigated through careful design, which places high demands on both experimental and numerical methodologies.
We propose a fully parametric geometrical model of an axial blood pump, designed to serve as a common benchmark for CFD studies. The model is defined by a set of geometric parameters, enabling both systematic parametric studies and automated optimization workflows. This generic yet realistic representation of a VAD facilitates consistent comparison of simulation results and supports reproducible research in this field.
To fully leverage this parametric model, we further present an extensible simulation workflow that integrates the parametric CAD generation with preprocessing, CFD simulation, postprocessing, and automated report creation. The entire framework was developed in-house and is designed to support systematic performance assessment and shape refinement, while being flexible and extensible. Our long term goal is to establish a reproducible and shareable benchmark that can support future extensions, including for example the integration of advanced metrics such as blood damage prediction and thrombus formation modeling.
Abstract: Transcatheter Aortic Valve Intervention (TAVI) is now the dominant form of aortic valve replacement over surgical alternatives, being performed over 200,000 times a year world-wide. However, leaflet and aortic thrombosis, though sub-clinical, is frequently observed following the procedure. There is a lack of consensus and conflicting clinical evidence on its eventual impact and mechanism on the patient outcome. Given the implants are not retrievable, and their indication is simultaneously expanding to younger patients, there is pressing urgency to understand how myriad of factors involved in TAVI combine to promote thrombosis development. In this talk, we present our ongoing computational analysis of haemodynamics in post-TAVI settings, integrating patient-specific anatomy, physiological measurements, and commercial prosthetic devices in fluid-structure interaction simulations.
Abstract:
Abstract: Cardiovascular diseases often lead to the deterioration or malfunction of native heart valves, disrupting normal blood circulation through the heart. In such cases, valve replacement becomes essential to restore proper cardiac function. Bileaflet mechanical heart valves are among the most widely used prosthetic devices due to their durability and favourable hemodynamic performance. Their two-leaflet design allows efficient blood flow while providing long-term structural reliability, making them a preferred choice for patients requiring permanent valve replacement. However, these valves can generate complex flow patterns and high shear stresses near the leaflets, which may promote thrombus (blood clot) formation. Consequently, patients typically require lifelong anticoagulation therapy, increasing the risk of bleeding complications. This talk explores whether incorporating superhydrophobic surfaces and vortex generators can further improve valve performance. Using computational fluid dynamics (CFD) simulations, the study models blood with its realistic non-Newtonian rheological behaviour under pulsatile flow conditions. Furthermore, blood is treated as a multiphase system, enabling detailed analysis of phenomena such as red blood cell (RBC) damage, which is a critical factor in designing safer and more efficient mechanical heart valves.
Abstract: Digital twins are emerging as a key paradigm in personalized medicine, enabling patient-specific simulations to support diagnosis and treatment planning. In the class of problems considered in our work, this requires solving partial differential equations with personalized parameters and geometries. The personalization procedure often requires repeated solutions of the governing equations; therefore, classical methods such as the finite element method can become too computationally expensive for clinical use. We present a surrogate modelling strategy designed to accelerate simulations on parametrized geometries. The method relies on the NN-PGD framework, combining physics-informed training with an interpretable architecture. In this talk, we introduce the main principles of the method and illustrate them on simplified model problems.
Abstract: Distinct features of the wall shear stress (WSS) topological skeleton, comprising fixed points and manifolds, have implications for near-wall transport and mechanobiology. Recently, spearheaded by my colleagues at Politecnico di Torino, we presented a framework linking the topologies of WSS and near-wall vorticity, and identifying allowed combinations of fixed points and their fluid mechanical implications. We have now shown that certain of these fixed-point combinations and topological structures allow to identify distinct vortical phenomena in the bulk flow that are analogous to tornadoes, downdrafts, and other atmospheric phenomena, each with potentially distinct mechanobiological implications.
In this presentation, I will review this WSS topological framework; present computational and magnetic resonance imaging evidence for tornadic phenomena in cerebral aneurysms; and discuss broader implications for characterization of hostile hemodynamics.
Abstract: This talk presents results from several years of research by our group on blood flow simulation in the aortic root. We present hemodynamic computations with an emphasis on robust and reliable numerical methods. We first study benchmark problems with analytical solutions and then turn to more realistic geometries, where full three-dimensional results are compared with axi-symmetric solutions. The main focus is on the effect of Navier slip at the vessel wall. We show that Navier slip can strongly influence the character of the flow and significantly change dissipation, wall shear stress and pressure drop, often more than fluid–structure interaction effects. The choice of boundary conditions becomes one of the key aspects of hemodynamic simulations.
Abstract: In this talk, we develop a fluid-structure-concentration interaction (FSCI) model including solid growth. Four solution variables are of interest: velocities and pressure of the liquid, solid displacements, and concentrations in the liquid and the solid. We formulate a monolithic system in Arbitrary Lagrangian Eulerian (ALE) coordinates. The resulting full order model is computationally analyzed in terms of mesh displacements, material parameter variations, mesh and time resolution using three different model setups. Specifically, the solid growth in the near-contact situation is a challenge for ALE and yields robust computational results in our cases. Based on those results we develop a reduced order model for the linearized FSCI system in order to reduce the computational cost. As before, our findings are analyzed for physical quantities of interest, computational cost, and comparisons to the full order model.
Abstract: Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Abstract: Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Abstract: Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.