Improving rupture status prediction for intracranial aneurysms using wall shear stress informatics

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Fluid Dynamics

Background:

• Wall shear stress (WSS) is critical in intracranial aneurysm (IA) development and progression.
• Previous studies have not fully utilized spatial WSS variations for predicting IA rupture.
• Accurate prediction of IA rupture status is crucial for clinical decision-making.

Purpose of the Study:

• To assess the feasibility of using spatial WSS patterns to predict IA rupture status.
• To introduce and validate a novel "WSS-informatics" approach for analyzing hemodynamic data.
• To improve the predictive modeling of IA rupture compared to conventional methods.

Main Methods:

• Patient-specific computational fluid dynamics (CFD) simulations were conducted for 112 IAs.
• WSS data from unstructured CFD meshes were mapped onto a unit disk, creating image-like data.
• Radiomics analysis was applied to extract spatial patterns from the mapped WSS data (WSS-informatics).

Main Results:

• Conventional WSS parameters did not correlate with IA rupture status.
• WSS-informatics metrics showed significant discriminative ability for IA rupture status (p < 0.05).
• Predictive models incorporating WSS-informatics features demonstrated improved performance (AUROC: 0.78 vs. 0.85, p < 0.01).

Conclusions:

• This study presents the first application of spatial WSS patterns, via WSS-informatics, to predict IA rupture.
• WSS-informatics offers a promising new method for enhancing the prediction of intracranial aneurysm rupture.
• The findings highlight the potential of integrating advanced data analysis techniques with CFD for improved clinical outcomes.