Real-World Variability in the Prediction of Intracranial Aneurysm Wall Shear Stress: The 2015 International Aneurysm CFD Challenge

Kristian Valen‐Sendstad ,Michael Barbour,Arjan J Geers,Hernán G Morales,Charles B L M Majoie,Masaaki Shojima,Sylvia Saalfield,Jan Bruening,Stephen P Broderick ,Jordi Pallarès ,Leonid Goubergrits,Kerem Pekkan,Adam Updegrove ,Motomu Tsuji ,Kuniyasu Niizuma,Ignacio Larrabide,Kenichi Kono,Salvatore Cito,Kartik Jain,Nicolas Aristokleous,Yahia M Al-Smadi,Neil Ashton,Prahlad G Menon,Thomas A Spirka ,Alistair Brown ,Philipp Berg,Alberto Aliseda,Bart Cornelissen ,Aslak W Bergersen,Shawn C Shadden,Shin Sugiyama ,Justin Tran ,Aike Qiao,Sherif Rashad,Alistair Revell,Fujimaro Ishida,Sreeja Vaippummadhom,Muhammad Owais Khan ,George Albert Einstein ,Simona Hodis,Kent‐André Mardal ,Yuji Shimogonya,Satoshi Koizumi,Şenol Pişkin ,Saikiran Rapaka,John G Houston ,Kristian Debus,Alison Marsden ,Sabine Roller,Thierry Léfèvre ,Viorel Mihalef,Leonard D Browne,Ramji Kamakoti,Wenyu Fu,David A Steinman

doi:10.1007/s13239-018-00374-2

Abstract

PurposeImage-based computational fluid dynamics (CFD) is widely used to predict intracranial aneurysm wall shear stress (WSS), particularly with the goal of improving rupture risk assessment. Nevertheless, concern has been expressed over the variability of predicted WSS and inconsistent associations with rupture. Previous challenges, and studies from individual groups, have focused on individual aspects of the image-based CFD pipeline. The aim of this Challenge was to quantify the total variability of the whole pipeline.Methods3D rotational angiography image volumes of five middle cerebral artery aneurysms were provided to participants, who were free to choose their segmentation methods, boundary conditions, and CFD solver and settings. Participants were asked to fill out a questionnaire about their solution strategies and experience with aneurysm CFD, and provide surface distributions of WSS magnitude, from which we objectively derived a variety of hemodynamic parameters.ResultsA total of 28 datasets were submitted, from 26 teams with varying levels of self-assessed experience. Wide variability of segmentations, CFD model extents, and inflow rates resulted in interquartile ranges of sac average WSS up to 56%, which reduced to < 30% after normalizing by parent artery WSS. Sac-maximum WSS and low shear area were more variable, while rank-ordering of cases by low or high shear showed only modest consensus among teams. Experience was not a significant predictor of variability.ConclusionsWide variability exists in the prediction of intracranial aneurysm WSS. While segmentation and CFD solver techniques may be difficult to standardize across groups, our findings suggest that some of the variability in image-based CFD could be reduced by establishing guidelines for model extents, inflow rates, and blood properties, and by encouraging the reporting of normalized hemodynamic parameters.

Highlights

Since the first individual case studies were published more than 15 years ago,[18,22,42] medical image-based computational fluid dynamics (CFD) of intracranial aneurysms has become a widely-used tool for elucidating the role of hemodynamic forces in aneurysm development and rupture.[39]
Wide variability exists in the prediction of intracranial aneurysm wall shear stress (WSS), irrespective of experience with image-based aneurysm CFD
This serves as an impediment to the integration of studies from different groups,[5] a step that may be required in order to achieve statistically significant findings in light of the many factors, other than hemodynamic forces, that influence aneurysm growth and rupture.[37]

Summary

Introduction

Since the first individual case studies were published more than 15 years ago,[18,22,42] medical image-based computational fluid dynamics (CFD) of intracranial aneurysms has become a widely-used tool for elucidating the role of hemodynamic forces in aneurysm development and rupture.[39]. Image-based CFD is subject to numerous sources of uncertainty along its pipeline: the clinical modality used to image the aneurysm[4,16,17]; digital segmentation of the lumen, often requiring subjective decisions about thresholds, filtering, smoothing, etc.[15,34,38]; truncation of the domain and attendant assumptions about velocity boundary conditions[7,19,30]; the need to assume flow rates,[21,25,32] since patient-specific measurements are rarely available; the pragmatic assumption of rigid walls 2,12,46 and simple blood rheologies[6,27,48] when, patient-specific properties are difficult or impossible to obtain; and the choice of mesh and time-step resolutions, as well as other CFD solver settings.[13,44,45] Common to the above-cited studies is that they were performed by individual research teams and focused on a single source of variability, all other factors being equal. But not perfect, agreement was found for cycle-averaged velocity patterns

Objectives

Methods

Results

Discussion

Conclusion