Graph Deep Learning for Intracranial Aneurysm Blood Flow Simulation and Risk Assessment

2025-12-09

AI Summary

Intracranial aneurysm rupture risk is strongly associated with local hemodynamics—particularly wall shear stress (WSS) and oscillatory shear index (OSI)—yet conventional computational fluid dynamics (CFD) simulations require hours and domain expertise, while 4D Flow MRI suffers from limited spatial resolution and high cost. To address this, we propose a novel graph neural network surrogate model that uniquely integrates a graph Transformer with an autoregressive prediction architecture. Given only a patient-specific vascular geometry graph, the model infers full-field WSS and OSI in seconds. It achieves zero-shot generalization across unseen anatomies and inflow conditions without calibration, completes inference for a single cardiac cycle in under one minute, and matches high-fidelity CFD accuracy. The model has been integrated into clinical imaging workflows, enabling bedside, high-resolution, real-time rupture risk assessment without reliance on CFD specialists.

Technology Category

• Computer Vision: 3D Computer Vision
• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Computational social science
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Intracranial aneurysms remain a major cause of neurological morbidity and mortality worldwide, where rupture risk is tightly coupled to local hemodynamics particularly wall shear stress and oscillatory shear index. Conventional computational fluid dynamics simulations provide accurate insights but are prohibitively slow and require specialized expertise. Clinical imaging alternatives such as 4D Flow MRI offer direct in-vivo measurements, yet their spatial resolution remains insufficient to capture the fine-scale shear patterns that drive endothelial remodeling and rupture risk while being extremely impractical and expensive. We present a graph neural network surrogate model that bridges this gap by reproducing full-field hemodynamics directly from vascular geometries in less than one minute per cardiac cycle. Trained on a comprehensive dataset of high-fidelity simulations of patient-specific aneurysms, our architecture combines graph transformers with autoregressive predictions to accurately simulate blood flow, wall shear stress, and oscillatory shear index. The model generalizes across unseen patient geometries and inflow conditions without mesh-specific calibration. Beyond accelerating simulation, our framework establishes the foundation for clinically interpretable hemodynamic prediction. By enabling near real-time inference integrated with existing imaging pipelines, it allows direct comparison with hospital phase-diagram assessments and extends them with physically grounded, high-resolution flow fields. This work transforms high-fidelity simulations from an expert-only research tool into a deployable, data-driven decision support system. Our full pipeline delivers high-resolution hemodynamic predictions within minutes of patient imaging, without requiring computational specialists, marking a step-change toward real-time, bedside aneurysm analysis.

Problem

• Simulates intracranial aneurysm hemodynamics quickly using graph neural networks
• Replaces slow CFD and low-resolution MRI for clinical risk assessment
• Enables real-time, high-resolution flow predictions for bedside decision support

Innovation

• Graph neural network surrogate model for rapid hemodynamics simulation
• Combines graph transformers with autoregressive predictions for accuracy
• Enables near real-time inference integrated with clinical imaging pipelines