Keynote I
chairperson: Alfons Hoekstra
Nic Smith
Multi-scale Computational Modelling of Coronary Blood Flow
Dutch Heart Foundation Lecture
Abstract
Jack Lee, Matt Sinclair, Eoin Hyde, Andrew Cookson, Simone Rivolo, Radomir Chabiniok, David Nordsletten, Eike Nagel, Maria Seibes, Jos Spaan, Nic Smith
The complexities associated with understanding coronary blood flow are in many cases produced by the number of determinants that control its function including network anatomy, systemic afterload and mechanical interaction with the myocardium throughout the cardiac cycle. Furthermore, the location of coronary artery disease is often distributed heterogeneously. In large epicardial vessels, coronary disease status can often be determined directly from anatomically based angiograms. However, in contrast, small vessel and microcirculatory dysfunction must be inferred indirectly from functional measurements including experimental quantification using microspheres or in the clinic MR first pass perfusion and imaging of myocardial motion. This range of spatial scales combined with the requirement to integrate multiple data types to analyse coronary perfusion highlights the need for a validated multi-scale framework that captures the relevant details at each level of the network. To address this need in this study we present our integrated coronary blood flow model with the capacity to span detailed anatomically based coronary flow modelling integrated with frameworks which represent microvascular perfusion and cardiac contraction. Specifically large deformation finite element models of contraction are coupled with reduced one-dimensional representations of vascular flow to simulate coronary perfusion. To enable direct comparison with both experimental measurements and clinically relevant imaging both microsphere and contrast agent transport are simulated respectively. Applying this approach we investigate and analyse experimental and clinical data sets focusing on enhancing the insight provided by functional data using computational models. The results of these simulations show that experimental microsphere deposition is systematically effected by network morphology and spatial scale. Additionally and of clinical relevance we show that the diffusive properties of contrast agents have a significant effect on the identification of perfusion defects in the clinic.