HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Simulation numérique de déploiement de stent coronaire en vue d’un traitement personnalisé – validation sur fantôme acquis en imagerie microdensitométrique

Abstract : The research work presented here takes its roots in the medical context of an interventional treatment for coronary artery disease, which consists in deploying an endovascular metallic prosthesis - called stent - inside a pathological artery. As the intervention takes place entirely inside the vessels, making sure that the stent placement is successful (i.e. unlikely to cause complications) can be a difficult task, particularly in some scenarios where the artery is complex to treat and requires adaptation from the cardiologist during stent apposition.The main objective of this PhD dissertation is to develop a method for numerically simulating the stent deployment inside a coronary artery, using a realistic model of mechanical behaviour. The simulation aims at guiding physicians during the apposition procedure, to be, in the long run, integrated into clinical routine. This purpose strongly constrains the simulation method, which has to be computationally efficient in order to not slow down significantly the intervention, while sufficiently precise for clinical use. The present thesis thus describes a numerical method to simulate the deployment of a balloon-expandable stent inside a coronary artery. Its main contributions can be divided into three distinctive parts.At first, we detail a generic method to represent a stent as a mesh of connected beam elements, suitable for finite element simulation. This method can be applied to any stent with a common design (i.e. made of rings and connectors), and easily adapted to any length or diameter. The use of beam elements allows to take advantage of the wire-like structure of the stent, reducing the number of degrees of freedom and lowering computational time.Beam elements are then associated with a model of plastic mechanical behaviour. To correctly simulate the deformation of the stent metallic alloys, we use a Von Mises-Hill plasticity model with mixed (kinematic and isotropic) linear hardening. Solution of plastic equations relies on a radial return algorithm. We also approximate the balloon inflation by connecting the stent mesh to a virtual balloon membrane using stiff springs. This additional contribution allows to stabilise the simulation, while taking into account the compliance tables provided by the stent manufacturers.Complementing the numerical model, we finally present an experimental validation protocol for the simulation. This protocol relies on the design and manufacturing of coronary artery phantoms, used for the deployment of ten commercial stents under micro-computed tomography (micro-CT) supervision. Each deployment is reproduced numerically using the simulation, and the final geometry of the simulated stent is compared to the micro-CT acquisition of the real stent, using rigid registration. The results we obtained show that the simulation precision has the same order of magnitude as the stent strut thickness (around 100 µm).This thesis is a first proof of concept for a realistic stent deployment simulation, which aims at being integrated into clinical routine. From the stent mechanical behaviour point of view, we present a method which is both accurate (experimentally validated), and faster than classic simulation methods from literature (using 3D elements). Two main leads are currently investigated to move towards a concrete clinical application. The first one is the optimisation of simulation features for which the computational time remains prohibitive (especially collision handling). The second one is the development of patient-specific artery models, retrieved from optical coherence tomography imaging data.
Complete list of metadata

Contributor : Abes Star :  Contact
Submitted on : Monday, January 10, 2022 - 9:35:12 AM
Last modification on : Wednesday, January 12, 2022 - 3:47:59 AM
Long-term archiving on: : Monday, April 11, 2022 - 6:52:26 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03518646, version 1


Camille Krewcun. Simulation numérique de déploiement de stent coronaire en vue d’un traitement personnalisé – validation sur fantôme acquis en imagerie microdensitométrique. Bio-informatique [q-bio.QM]. Université Clermont Auvergne [2017-2020], 2020. Français. ⟨NNT : 2020CLFAC086⟩. ⟨tel-03518646⟩



Record views


Files downloads