Virtual Reality and AI for Real-Time Clinical Decision Support and Intracranial Aneurysm Risk Prediction

Objective

Intracranial aneurysms (IAs) pose a severe neurovascular threat, with rupture leading toIntracranial aneurysms (IAs) pose a severe neurovascular threat, with rupture leading to life-threatening complications. Current treatment planning relies on static imaging and computational fluid dynamics, which are computationally expensive and lack real-time clinical integration. Clinicians face uncertainty in selecting optimal intervention strategies, increasing procedural risks and costs. VRAI-CURE proposes an AI-driven clinical decision support system that combines Physics-Constrained Graph Neural Networks for real-time hemodynamic simulation and immersive Virtual Reality (VR) visualization. This system enables instantaneous, patient-specific blood flow predictions, transforming clinical workflows by providing neurovascular specialists with actionable insights for treatment planning. Unlike conventional simulation or imaging methods, the proposed AI-based approach is computationally efficient, self-improving with each new patient case, and seamlessly integrates into standard imaging workflows (MRI, CT, angiography). By delivering an interactive VR-based visualization of blood flow dynamics, VRAI-CURE enhances clinical understanding of aneurysm behavior and treatment effects, facilitating personalized decision-making. Our preliminary validation demonstrates superior accuracy and efficiency compared to existing techniques, addressing a significant gap in neurovascular intervention. With the VR healthcare market projected to reach $40 billion by 2032, the transformative potential extends beyond neurovascular medicine, offering a cutting-edge solution in a rapidly expanding sector. This project aims to advance to TRL 6, developing a certified clinical prototype ready for industrial transfer. The project’s high-risk/high-gain potential positions it as a transformative innovation in precision neurovascular medicine, with broad implications for AI-driven clinical decision support.