Clustering Algorithms
An inverse problem approach to the estimation of volume change
MICCAI'05 Proceedings of the 8th international conference on Medical image computing and computer-assisted intervention - Volume Part II
Hi-index | 0.00 |
Weak spots in the aneurysm could be identified estimating the regional stiffness of the wall. Our approach consists in defining a parametric biomechanical model of the vessel which, given the patient's vascular morphology and the blood in- and outflow obtained from non-invasive imaging as well as parameters describing the local elasticity of the wall, enables the computation of the theoretical deformed wall position. The distance between this latter and the one obtained from the aneurysm pulsation is iteratively minimized in order to estimate the optimal set of stiffness parameters. In order to reduce the number of variables to estimate, the aneurysm morphology is clustered into a limited number of regions with uniform stiffness. A random noise perturbation (