Research tools

Our computer models range from so-called lumped parameter models (0D) of an organ or system (e.g. the interaction of cardiac assist devices with the native heart) to complex 3D numerical models of artificial organs (artificial kidney, artificial liver), the heart, blood vessels (the carotid artery, aortic aneurysms), medical devices (heart valves, stents), or in conditions where these domains meet (e.g. in the study of the vascular access site for hemodialysis, or flow in catheters positioned in blood vessels or heart cavities). In many cases, the geometries for the simulations are obtained from medical images acquired with (micro-)CT, MRI or ultrasound. Computations are done using Fluent (fluid mechanics solver) and Abaqus (structure mechanics solver) or both, accounting for the interaction between the fluid and structure domain (fluid-structure interaction). Fluid-structure interaction is, at present, one of our key research domains.

More and more, advanced computer simulations become a substantial part of the design and testing of new devices. A domain where we are particularly making use of these new virtual testing modalities is in the design of stents and stent-related devices. Parametric modelling of these devices, making use of the extensive possibilities of pyFormex , allows to generate hundreds of variants of a particular design of a device in an matter of seconds, and direct linking of pyFormex with Finite Element Software packages like Abaqus provides the means to virtually subject these variants to mechanical tests to assess, for instance, their radial strength or deformation characteristics. These simulations become more and more realistic, taking into account more and more aspects of the real-world situation, such as - for our stent research - the unfolding of the balloon, and the complexities of the geometry (stenting of bifurcations) and the (pathological) vessel wall.

Irrespective of the tremendous potential of computer based design and testing, computer models still require validation and the final design of a device is also best subjected to testing in the real, physical world. Experimental testing thus remains a cornerstone of our research. From the collection of experimental set-ups and hydraulic bench models developed over the last decade, our 1:1 scale model of the human left heart and the arterial tree is still one of our showpieces. It allows us to study fundamental aspects of cardiac (systolic and diastolic) function and cardiovascular hemodynamics and patho-physiology (e.g. arterial wave reflection, blood pressure waveform analysis, arterial mechanics) as well as applied research on prosthetic heart valves, cardiac assist devices, endovascular applications, etc... The model allows high-fidelity measurements on pressures, ventricular volume and flow; it is MRI compatible and has access windows for ultrasound-based visualization of the mitral and aortic valve from a trans-thoracic and trans-oesophageal window. Other hydraulic bench work includes testing of hemodialyzers, artificial lungs, coronary bypass grafts, the brain vasculature, etc...

Here also, we strive towards the simulation of realistic patient-specific geometries and settings, with models being home-made by our technical staff, eventually based on models generated via rapid-prototyping techniques. When detailed flow velocity and flow field visualizations are desired (e.g. to validate and/or confirm numerical results), we produce (enlarged) models, accounting for the physical laws of dynamic similarity, and visualize the flow using Particle Image Velocimetry (PIV).

Finally, our group is also a partner in applied clinical or pre-clinical research at our own university hospital and abroad, where we directly cooperate with the physicians, dealing with the technical and mechanical aspects of the research. We specifically refer here to the "Asklepios" study, where over 2500 inhabitants of Erpe-Mere, aged between 35 and 55 years old, were screened in 2003 and 2004 for cardiovascular disease using advanced ultrasound techniques and applanation tonometry (in addition to biological and biochemical analysis). In collaboration with the cardiology and bio-engineering department of our university, we strive towards improved quantification of cardiovascular structure and function, with the long-term goal of developing markers allowing early detection of atherosclerosis. Other pre-clinical research focuses on quantification of mechanical wall stress and fluid shear stress in different vascular territories, both from the perspective of their role as mediators of atherosclerosis (carotid bifurcation), as triggers of biological responses to flow and stress disturbances (flow-mediated dilatation protocols; vascular remodelling), and as potential co-factors in embryonic heart development and in genetically determined diseases such as Marfan syndrome.