Annals of Biomedical Engineering

, Volume 39, Issue 2, pp 884–896

Computational Hemodynamics in Cerebral Aneurysms: The Effects of Modeled Versus Measured Boundary Conditions

  • Alberto Marzo
  • Pankaj Singh
  • Ignacio Larrabide
  • Alessandro Radaelli
  • Stuart Coley
  • Matt Gwilliam
  • Iain D. Wilkinson
  • Patricia Lawford
  • Philippe Reymond
  • Umang Patel
  • Alejandro Frangi
  • D. Rod Hose
Article

DOI: 10.1007/s10439-010-0187-z

Cite this article as:
Marzo, A., Singh, P., Larrabide, I. et al. Ann Biomed Eng (2011) 39: 884. doi:10.1007/s10439-010-0187-z

Abstract

Modeling of flow in intracranial aneurysms (IAs) requires flow information at the model boundaries. In absence of patient-specific measurements, typical or modeled boundary conditions (BCs) are often used. This study investigates the effects of modeled versus patient-specific BCs on modeled hemodynamics within IAs. Computational fluid dynamics (CFD) models of five IAs were reconstructed from three-dimensional rotational angiography (3DRA). BCs were applied using in turn patient-specific phase-contrast-MR (pc-MR) measurements, a 1D-circulation model, and a physiologically coherent method based on local WSS at inlets. The Navier–Stokes equations were solved using the Ansys®-CFX™ software. Wall shear stress (WSS), oscillatory shear index (OSI), and other hemodynamic indices were computed. Differences in the values obtained with the three methods were analyzed using boxplot diagrams. Qualitative similarities were observed in the flow fields obtained with the three approaches. The quantitative comparison showed smaller discrepancies between pc-MR and 1D-model data, than those observed between pc-MR and WSS-scaled data. Discrepancies were reduced when indices were normalized to mean hemodynamic aneurysmal data. The strong similarities observed for the three BCs models suggest that vessel and aneurysm geometry have the strongest influence on aneurysmal hemodynamics. In absence of patient-specific BCs, a distributed circulation model may represent the best option when CFD is used for large cohort studies.

Keywords

Computational fluid dynamicsPhase-contrast MRI1D circulation model

Copyright information

© Biomedical Engineering Society 2010

Authors and Affiliations

  • Alberto Marzo
    • 1
  • Pankaj Singh
    • 2
  • Ignacio Larrabide
    • 3
    • 4
    • 5
  • Alessandro Radaelli
    • 6
  • Stuart Coley
    • 7
  • Matt Gwilliam
    • 8
  • Iain D. Wilkinson
    • 9
  • Patricia Lawford
    • 1
  • Philippe Reymond
    • 10
  • Umang Patel
    • 11
  • Alejandro Frangi
    • 3
    • 4
    • 5
  • D. Rod Hose
    • 1
  1. 1.Department of Cardiovascular Science, Academic Unit of Medical Physics, Faculty of Medicine and Biomedical SciencesUniversity of SheffieldSheffieldUK
  2. 2.Departments of Medical Physics and NeurosurgeryRoyal Hallamshire HospitalSheffieldUK
  3. 3.Centre for Computational Imaging and Simulation Technologies in BiomedicineUniversitat Pompeu Fabra (UPF)BarcelonaSpain
  4. 4.Centre for Networked Biomedical Research on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN)BarcelonaSpain
  5. 5.Institució Catalana de Recerca i Estudis Avaçats (ICREA)BarcelonaSpain
  6. 6.Philips HealthcareBestThe Netherlands
  7. 7.Department of NeuroradiologyRoyal Hallamshire HospitalSheffieldUK
  8. 8.Medical Physics and Clinical EngineeringSheffield Teaching Hospitals NHSSheffieldUK
  9. 9.Department of Human Metabolism, Academic Unit of RadiologyUniversity of SheffieldSheffieldUK
  10. 10.Laboratory of Hemodynamics and Cardiovascular Technology (LHCT)École Polytechnique Fédérale de LausanneLausanneSwitzerland
  11. 11.Department of NeurosurgeryRoyal Hallamshire HospitalSheffieldUK