Optimal MR Protocol for the monitoring of SVD patients at Low mAgnetic field – OPLA

Low-magnetic-field MRI imagers are attracting increasing interest due to reduced purchase and maintenance costs, a minimized carbon footprint, and reduced space requirements. However, two issues limit their expansion: low signal-to-noise ratio (SNR) and unconventional contrasts. Thus, these systems are currently unattractive, particularly for the detection and monitoring of fine brain abnormalities, such as those that develop in patients with small vessel disease (SVD). Indeed, this pathology creates brain abnormalities, including enlarged perivascular spaces, white matter hyperintensities, and microbleeds. These abnormalities are detectable by high-resolution MRI at high magnetic field by applying different specific sequences.
Our project therefore aims to develop a fast and robust MRI protocol to detect these three brain abnormalities in patients with SVD at low magnetic field, here 0.55T. To achieve this, our expertise in exotic Fourier space encoding will facilitate the production of 3D images with high SNR and adjustable contrasts. Thus, three sequences will be developed, not only to determine the optimal parameters for detecting the three brain anomalies mentioned above, but also to develop Artificial Intelligence (AI) algorithms to minimize acquisition times while improving the signal and spatial resolution of the images. A first deliverable is therefore to develop a neuroimaging protocol dedicated to the detection of the cerebral consequences of SVD in 25 minutes. In parallel, automatic segmentation tools will be developed to be adapted to the new image contrasts obtained at 0.55T. These tools will be included in an open-source computerized container to provide the community with an automatic quantification tool. The robustness of our strategy will be validated on a cohort of 500 volunteers with SVD. The goal of the project is therefore to provide the scientific community with a pipeline, from image acquisition to analysis.
The project involves a multidisciplinary consortium bringing together experts in the fields of MRI physics, applied mathematics, and neuroimaging, as well as a company developing open-source automatic image segmentation tools.
The advances hoped for in this project will make SVD diagnosis accessible to at-risk populations, including those located in medical deserts and developing countries. This project is the first step in a long-term strategy aimed at bringing MRI machines closer to the patient (thanks to ultra-low-field machines currently under development) and even to populations, thereby expanding the scope of MRI applications to prevention. The pipeline will also benefit other pathologies such as multiple sclerosis and Alzheimer's disease.