MINING DATA FROM HEMODYNAMIC SIMULATIONS VIA MULTILAYER PERCEPTRON NEURAL NETWORK
M. Radović and N. Filipović Abstract
Arterial geometry variability is present both within and across individuals. A multilayer
perceptron neural network was proposed for mining data generated from computer simulations.
The proposed approach was applied to analyze the influence of geometric parameters on
maximal wall shear stress (MWSS) in the human carotid artery bifurcation. A parametric model
was used for generating a set of observational data that contains the maximum wall shear stress
values for a range of probable arterial geometries. The data set was mined via a multilayer
perceptron making it possible to predict values of the maximum wall shear stress for new
geometries that were not in the training data set.
