Research Papers

Finite Element Stent Modeling for the Postoperative Analysis of Transcatheter Aortic Valve Implantation

[+] Author and Article Information
Raoul Hopf

Institute of Mechanical Systems,
Department of Mechanical Engineering,
ETH Zurich,
Zurich 8012, Switzerland
e-mail: rhopf@ethz.ch

Michael Gessat

Hybrid Laboratory for Cardiovascular
Division of Cardiovascular Surgery,
University Hospital Zurich,
Zurich 8091, Switzerland;
Computer Vision Laboratory,
ETH Zurich,
Zurich 8092, Switzerland

Christoph Russ

Computer Vision Laboratory,
ETH Zurich,
Zurich 8092, Switzerland

Simon H. Sündermann, Volkmar Falk

Klinik für Herz-Thorax-Gefässchirurgie,
Deutsches Herzzentrum Berlin,
Berlin 13353, Germany

Edoardo Mazza

Institute of Mechanical Systems,
Department of Mechanical Engineering,
ETH Zurich,
Zurich 8012, Switzerland;
Swiss Federal Laboratories
for Materials Testing and Research,
Dübendorf 8600, Switzerland

1Corresponding author.

Manuscript received June 23, 2016; final manuscript received February 22, 2017; published online May 3, 2017. Assoc. Editor: Marc Horner.

J. Med. Devices 11(2), 021002 (May 03, 2017) (7 pages) Paper No: MED-16-1245; doi: 10.1115/1.4036334 History: Received June 23, 2016; Revised February 22, 2017

In order to evaluate the performance of stents used in transcatheter aortic valve implantation (TAVI), finite element simulations are setup to reconstruct patient-specific contact forces between implant and its surrounding tissue. Previous work used structural beam elements to setup a numerical model of the CoreValve stent used in TAVI and developed a procedure for implementing kinematic boundary conditions from noisy computer tomography (CT) scanning data. This study evaluates element size selection and quantitatively investigates the choice of a linear elastic constitutive model for the Nitinol stent under physiological loading conditions. It is shown that this simplification leads to reliable results and enables a huge reduction in computation time. Further, the procedure used to compensate for noisy postoperative CT data is tested by adding artificial noise. It is concluded that for physiologically relevant loading ranges, the procedure yields convergent results and successfully eliminates the influence of the noise.

Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

Heat maps for three representative patient cases, showing different forms of typical force fields. High homogeneity and relatively low radial forces (left), globally high radial forces (middle), and a single peak (right). White corresponds to zero force and black represents 0.5 N nodal force.

Grahic Jump Location
Fig. 2

Each intersection point is uniquely addressed by an index pair (i, j), where i represents the ith layer and j represents the jth point in the layer

Grahic Jump Location
Fig. 3

Basic geometric feature definitions on the CoreValve: The stent can be divided into 11 layers. (a) Each layer holds 15 intersection points Pij. There are30physical strings, which impose a node connectivity pattern for the intersections(b).

Grahic Jump Location
Fig. 4

The main coordinate system is placed at the center of the stent, such that the z-direction coincides with the axis of the undeformed stent and the xy-plane contains the zeroth layer. For each intersection node, a local coordinate system is computed. The dashed blue vectors are used to compute the normal direction e2 (see figure online for color).

Grahic Jump Location
Fig. 5

Iterative application of boundary conditions. The left-hand side shows the direct application of kinematic boundary conditions, which is not possible in this case. In the middle, the setup used for the iterative method is depicted: The initial length of the force element d0=||Pij−P̃ij|||1  μm is very small as compared to the average applied displacement. The right-hand side illustrates the bilinear force law with a cut-off at D and stiffness values k1 and k2.

Grahic Jump Location
Fig. 6

A representative patient simulation over ten iterations. Initial high peaks in reaction force reduce quickly, while nodal drifts remain within the range of measurement uncertainty.

Grahic Jump Location
Fig. 7

Results of the convergence analysis and different grades of meshing illustrated on a single strut (bottom right)

Grahic Jump Location
Fig. 8

Comparison between linear elastic and fully superelastic material models. Bottom right shows a tensile load cycle for a linear elastic rod and a superelastic rod at room and body temperature.

Grahic Jump Location
Fig. 9

Left: Mean radial force per layer after one iteration. Center: Mean radial force per layer after the convergence criterion is reached and the corresponding the statistical descriptors (right).




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In