Reconstruction of Shape and its Effect on Flow in Arterial Conduits

The geometry of arterial conduits derived from in vivo image data is subject to acquisition and reconstruction errors. This results in a degree of uncertainty in defining the bounding geometry for a patient-specific anatomical conduit. In applying computational fluid dynamics to model the flow in specific anatomical configurations, the effect of the uncertainty in boundary definition should be considered, particularly if the objective is to extract quantitative measures of the local haemodynamics.

Taking an example of a bypass graft configuration, we examine the effects of image threshold, surface smoothing and semi-idealization on the modelled geometry and the resulting flow. Procedures for reconstruction from medical images are outlined and applied with different parameter values within the image uncertainty range to create alternative models from the same data set.

Methods to characterize the flow structure and wall shear stress (WSS) are introduced and used to provide quantitative comparison of the different haemodynamic environments associated with the varying model geometries. Comparable effects on the WSS distribution are found to occur with progressively increased surface smoothing and semi-idealization of the geometry by elliptical section fitting. Significant differences in WSS correspond to different threshold choices