I034 Mechanical modeling of in vivo human carotid arteries from non-invasive clinical data. application to normal subjects

Abstract

For mechanical modeling, in vivo data are relatively incomplete in comparison to in vitro results. However, identification of mechanical properties from human clinical data to compute wall stress fields can play an important role in understanding better pathological evolutions. Demonstrate the feasibility of material identification and stress computation from clinical data for normal subjects. In vivo human common carotid arteries (CCAs) were explored non-invasively for 16 normal subjects. During several cardiac cycles, medial diameter, intimal-medial thickness and blood pressure were measured by a high-resolution echotracking (Art. Lab®) and applanation tonometry (SphygmoCor®), respectively. To study the wall mechanical behavior, the CCA was assumed to be a 3D hollow cylinder subjected to dynamical intraluminal pressure and perivascular constraints. We also assumed that the arterial wall is made of hyperelastic, fibrous, and incompressible material with smooth muscle activity and residual stresses. We included wall mechanical contributions by microconstituents: an elastin-dominated matrix, collagen fibers, and vascular smooth muscle (VSM). We solved the in vivo boundary value problem semianalytically to compute the intraluminal pressure during a cardiac cycle. Minimizing the difference between computed and measured inner pressures over the cardiac cycle provided the identification of optimal model parameters employing a nonlinear regression. The fit-to-data gave very good results and was possible in all cases. There was a convergence of parameters for major constituents such as collagen, elastin and VSM tone. Age was correlated with collagen content and residual stresses. The predicted radial, circumferential, and axial stretches and stresses within the wall during the cardiac cycle were sensible. We were able to reproduce the evolution of inner blood pressure identifying experimentally unknown geometric and material parameters directly from in vivo human data, in order to compute wall stresses and stretches over a cardiac cycle. We can extend the proposed approach to pathological cases such as hypertension.