gms | German Medical Science

65. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC)

Deutsche Gesellschaft für Neurochirurgie (DGNC) e. V.

11. - 14. Mai 2014, Dresden

In vivo validation of CFD simulations in aneurysms

Meeting Abstract

  • Christian Doenitz - Klink und Poliklinik für Neurochirurgie, Universitätsklinikum Regensburg
  • Christoph Palm - Hochschule für angewandte Wissenschaften, Regensburg – Medical Image Computing (Re-MIC), Regensburg
  • Johannes Platz - Klink und Poliklinik für Neurochirurgie, Johann-Wolfgang-Goethe Universität Frankfurt
  • Volker Seifert - Klink und Poliklinik für Neurochirurgie, Johann-Wolfgang-Goethe Universität Frankfurt
  • Alexander Brawanski - Klink und Poliklinik für Neurochirurgie, Universitätsklinikum Regensburg

Deutsche Gesellschaft für Neurochirurgie. 65. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC). Dresden, 11.-14.05.2014. Düsseldorf: German Medical Science GMS Publishing House; 2014. DocP 180

doi: 10.3205/14dgnc574, urn:nbn:de:0183-14dgnc5749

Veröffentlicht: 13. Mai 2014

© 2014 Doenitz et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Objective: Computational fluid dynamics (CFD) is increasingly being used for modelling hemodynamics in intracranial aneurysms. These simulations are used to study the formation, growth and rupture of cerebral aneurysms and may contribute to assess the risk of rupture and to improve the treatment. While CFD techniques are well established, the need for validation of simulated results remains. We used intraoperative indocyanine green video angiography (ICG) in aneurysm surgery to directly visualize flow patterns and compared them to the CFD results of these aneurysms.

Method: Intraoperative Indocyanine green (ICG) video angiography during surgery of middle cerebral artery aneurysms were screened for visible flow patterns. Images of these sequences were extracted frame-by-frame using conventional video capture software. The frames were analyzed with the ImageJ software v 1.47 ( To visualize the flow patterns we used colour-coded point of time of maximum gradient using the HSV colour model. 3D models of the middle cerebral artery aneurysms were reconstructed from preoperative 3D angiography using the current version of AneuFuse software v 2.2 (Supercomputing Solutions, Bologna, Italy). The solvers used within AneuFuse were ANSYS-ICEM and ANSYS-CFX v 14.5 (Ansys, Inc., Canonsburg, PA, USA). For post-processing visualization of the results and comparison to the ICGvideo analyses we used Avizo Wind 3D analysis software v 7.1.1 (Visualization Sciences Group SAS, USA).

Results: ICG video angiography revealed meaningful flow patterns in five of eight cases. These flow patterns could be visualized in a single image by analyzing the maximum gradient of brightness in all of these five cases. The streamline patterns of the CFD simulations correlated well with the flow seen in the ICG video angiography in all cases. The main streamlines, vortices and impingement points could be identified depending on the intraoperative angle of view.

Conclusions: Validation of computational fluid dynamics is crucial for the scientific value and acceptance of flow simulations and their impact on research of aneurysm biodynamics. We present a new tool for direct observation of in vivo flow patterns that can help to understand and confirm the influence of various variables on CFD simulations.