Vanguard Low-Field MRI for Acute Stroke diagnosis – VLF-Stroke

Stroke, 85% of which is ischemic, is the leading cause of acquired disability in adults and one of the most time-critical conditions in medicine. High-Field MRI (HF-MRI) is the gold standard for acute stroke diagnosis in France, but its use is limited by infrastructure requirements, high costs, and accessibility challenges, particularly in underserved areas. Low-Field MRI (LF-MRI), operating below 1 Tesla, presents a promising alternative due to its portability, lower cost, and lower infrastructure requirements. However, its clinical utility is currently hindered by longer acquisition times and lower image quality compared to HF-MRI. Moreover, so far, no published study has prospectively assessed LF-MRI in the target population of patients with suspected hyper-acute stroke.
This project aims to overcome these limitations by optimizing LF-MRI for acute stroke management through a comprehensive research strategy. Initially, we will collect data from healthy volunteers and subacute stroke patients to optimize acquisition parameters and reduce acquisition time. This data will support the development of advanced AI-based imaging techniques, including image-to-image denoisers, plug-and-play reconstruction frameworks, and generative models, designed to enhance image quality and reduce MRI acquisition times.
We will then establish a cohort of patients with suspected hyperacute stroke and explored by HF-MRI, who will undergo LF-MRI examination as well. This will allow us to assess the diagnostic accuracy of LF-MRI compared to HF-MRI in a real-world clinical setting. Simulated treatment decision-making will be performed to evaluate whether LF-MRI can be safely used for treatment decisions in this context.
The project will also build a health economic model to compare the cost-effectiveness of LF-MRI against HF-MRI, providing valuable insights into the potential economic benefits of integrating LF-MRI into clinical practice. This economic analysis will inform the development of a roadmap toward a phase 3 clinical trial in a hospital setting, comparing standard care with LF-MRI-enabled care.
Finally, we will explore the feasibility of integrating LF-MRI into Mobile Stroke Units, which are currently equipped with CT scanners. This integration could leverage the higher sensitivity of MRI in the prehospital setting, facilitating fast diagnosis and treatment initiation.
This project represents a significant leap forward in stroke imaging technology and offers substantial translational value by bridging the gap between early-stage research and clinical application. By uniting a diverse consortium of partners, including academic institutions, healthcare providers, and computer scientists specialized in artificial intelligence, we aim to lay the groundwork for future phase 3 clinical trials. Our goal is to expand access to high-quality stroke diagnostics, particularly in remote and underserved settings, ultimately enhancing stroke care outcomes and improving patient prognosis.