AI MODELS OF THE HEMODYNAMIC SIMULATION
 
O. Miljković, M. Ivanović, N. Filipović, M. Kojić
Abstract
A trial to find the right reasons of the pathogenesis of the pathological changes on the blood
vessels and to prevent the growing of the cardiovascular diseases, includes many different
techniques of the Artificial Intelligence (AI). Fuzzy modeling and Neural Network reasoning
are shown in this paper as two possible solutions in measuring valuable hemodynamic factors in
the arterial blood circulation. The interactive program MedCFD, using Finite Element Method
(FEM), provides automatic generation of a realistic geometric model of an aneurysm, as the
most frequent pathological change on the blood vessels, and offers possibility to determine and
evaluate the effects of flow characteristics. Its inputs and outputs will be training data for neural
network model and a source of the knowledge for fuzzy modeling of the same real system.
Input parameters of these models are geometrical parameters of the aneurysm. The values of the
maximum shear stress at inlet and outlet zones are their output parameters. Both of artificial
intelligence models approximate the MedCFD calculation of these important hemodynamic
factors and are able to simulate and give good responses to new inputs. A significant
achievement of this approach is its time saving feature. Instead of solving large systems of
equations which may last for days, we are able to collect results of fuzzy or neural network
model of the same phenomena instantaneously.