Dissemin is shutting down on January 1st, 2025

Published in

Elsevier, Ultrasound in Medicine and Biology, 6(30), p. 803-813

DOI: 10.1016/j.ultrasmedbio.2004.04.005

Links

Tools

Export citation

Search in Google Scholar

A finite element model for performing intravascular ultrasound elastography of human atherosclerotic coronary arteries.

This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Red circle
Postprint: archiving forbidden
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

Intravascular ultrasound (US) elastography measures in an artery the arterial radial strain and displays it in an elastogram. An elastogram adds diagnostic information, such as the proneness of a plaque to rupture and its material composition. However, radial strain depends upon the material properties of an artery, including geometry and used catheter position. Therefore, there is not always a one-to-one correspondence between radial strain and rupture-proneness or material composition. Both the dependence and the correspondence can be quantified after a proper finite element model (FEM) is available. Therefore, this paper proposes a FEM and shows that it can model the arterial strain behavior. Its modelling capability was evaluated by comparing simulated with measured elastograms. Measured elastograms were processed from radiofrequency (RF) data obtained in vitro from six objects: a vessel-mimicking phantom and five excised human atherosclerotic coronary arteries. A FEM was created for each object and used to simulate an elastogram; the material properties and geometry of the FEM were obtained from the histology of the object. Comparison was performed upon high strain regions (HStR), because these regions have proven to contain plaques that show the hallmarks of vulnerable plaques. Eight HStR were automatically identified from the five arteries. Statistical tests showed that there was no significant difference between simulated and corresponding measured elastograms in location, surface area or mean strain value of a HStR. The results demonstrate that the FEM can simulate elastograms measured from arteries. As such, the FEM may help in quantifying strain-dependencies and assist in tissue characterization by reconstructing a Young's modulus image from a measured elastogram.