Construção de modelos de árvores arteriais considerando o efeito Fahraeus-Lindqvist

Abstract

Arterial tree models have been successfully used to gain a better understanding of all hemodynamics aspects of clinically relevant regions of the human body, including diagnosis and applications in surgical planning. The main motivation for the construction of these models is the difficulty to obtain sufficient anatomical data to describe in detail the geometrical and topological structures of peripheral arterial networks. Basically, the models can be classified into: anatomical, lumped parameter, fractal and optimized. This work focuses on the generation of optimized models based on Constructive Constrained Optimization (CCO) method. CCO is capable of generating arterial tree models that reproduce characteristics of real coronary tree, such as pressure profiles, vessel diameter and bifurcation angle distribution. However, this method does not consider a realistic blood viscosity during the generation of models, i.e., it disregards the F˚ahraeus-Lindqvist effect, which indicates that the blood viscosity depends nonlinearly on diameter of the vessel in which blood is draining and on discharge of hematocrit. In this context, the work investigates an algorithm that takes into account this effect during the construction of models of arterial trees. Several scenarios of 2D/3D simulations using this algorithm were done in order to study the influence of the choice of blood viscosity on morphometric and hemodynamic properties of the models. The results indicate that the blood viscosity affects the distribution of vessel radii, the architecture and pressure profiles of the models generated through computer simulations. Furthermore, these in silico models are consistent with real coronary arterial trees.