Accuracy and Reproducibility of Patient-Specific Hemodynamic Models of Stented Intracranial Aneurysms: Report on the Virtual Intracranial Stenting Challenge 2011

S. Cito, A. J. Geers, M. P. Arroyo, V. R. Palero, J. Pallarés, A. Vernet, J. Blasco, L. San Román, W. Fu, A. Qiao, G. Janiga, Y. Miura, Makoto Ohta, M. Mendina, G. Usera, A. F. Frangi

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)

Abstract

Validation studies are prerequisites for computational fluid dynamics (CFD) simulations to be accepted as part of clinical decision-making. This paper reports on the 2011 edition of the Virtual Intracranial Stenting Challenge. The challenge aimed to assess the reproducibility with which research groups can simulate the velocity field in an intracranial aneurysm, both untreated and treated with five different configurations of high-porosity stents. Particle imaging velocimetry (PIV) measurements were obtained to validate the untreated velocity field. Six participants, totaling three CFD solvers, were provided with surface meshes of the vascular geometry and the deployed stent geometries, and flow rate boundary conditions for all inlets and outlets. As output, they were invited to submit an abstract to the 8th International Interdisciplinary Cerebrovascular Symposium 2011 (ICS’11), outlining their methods and giving their interpretation of the performance of each stent configuration. After the challenge, all CFD solutions were collected and analyzed. To quantitatively analyze the data, we calculated the root-mean-square error (RMSE) over uniformly distributed nodes on a plane slicing the main flow jet along its axis and normalized it with the maximum velocity on the slice of the untreated case (NRMSE). Good agreement was found between CFD and PIV with a NRMSE of 7.28%. Excellent agreement was found between CFD solutions, both untreated and treated. The maximum difference between any two groups (along a line perpendicular to the main flow jet) was 4.0 mm/s, i.e. 4.1% of the maximum velocity of the untreated case, and the average NRMSE was 0.47% (range 0.28–1.03%). In conclusion, given geometry and flow rates, research groups can accurately simulate the velocity field inside an intracranial aneurysm—as assessed by comparison with in vitro measurements—and find excellent agreement on the hemodynamic effect of different stent configurations.

Original languageEnglish
Pages (from-to)154-167
Number of pages14
JournalAnnals of Biomedical Engineering
Volume43
Issue number1
DOIs
Publication statusPublished - 2014 Jan 1

Keywords

  • Challenge
  • Computational fluid dynamics
  • Intracranial aneurysm
  • Particle imaging velocimetry
  • Stents

ASJC Scopus subject areas

  • Biomedical Engineering

Fingerprint Dive into the research topics of 'Accuracy and Reproducibility of Patient-Specific Hemodynamic Models of Stented Intracranial Aneurysms: Report on the Virtual Intracranial Stenting Challenge 2011'. Together they form a unique fingerprint.

Cite this