Building an MRI data-lake: the potential of individual muscle quantification and data aggregation to refine biomarker understandings in FSHD

Investigators: Seth D Friedman PhD, Doris G Leung MD PhD

Category: Research - Basic

Put simply, 99.9% of the world's natural history MRI data collected in FSHD patients to date remains quantitatively unanalyzed at the individual muscle level. Overcoming this analytic bottleneck has created a new opportunity, that of combining data across sites into a data “lake” to rapidly advance what can be known about the natural history of the disease and better inform how to plan, conduct, and interpret drug trials. To make concerted steps toward this end goal, this project aims to add three diverse datasets to pilot data already analyzed with a novel AI approach. Aggregated data will help to fill in methodological gaps and test new hypotheses about the progression of fat infiltration over time. The second component of the project will focus on building relationships, workflows, and processes to allow for scaling of the project to include data from centers around the world and pursuing sustainable funding.