Fast acquisition techniques for diffusion tensor magnetic resonance imaging of the human heart in-vivo – FADIHEART

Heart diseases remain one of the most serious health problems and the leading cause of death in most countries in the western world, including France and China, despite decades of intensive cardio-vascular research in basic and clinical sciences. However, almost nothing is known about the intrinsically three-dimensional (3D) fiber structure of the human heart, and even less in-vivo, which is fundamental for a comprehensive understanding of the interrelations between mechanical function, hemodynamics, and adaptive structural changes in cardiac diseases. The only reason for this situation is that there is currently no means to access such 3D fiber structure in-vivo. The challenge of this project is to develop fast diffusion tensor magnetic resonance imaging (DT-MRI) for investigating the 3D fiber structure of the human heart in-vivo, using a global approach based on strong multidisciplinary interactions between MRI physicists, computer scientists, and clinicians. The approach consists in considering the fast imaging problem neither as a simple MRI sequence design problem, nor a simple image reconstruction problem, but as a whole in which required specific targets are achieved in a combined way. By designing new DT-MRI sequences through modelling and quantifying the influence of motion on signal loss, developing novel image reconstruction methods based on using the 2D singular monotonic function analysis model, proposing systematic mathematical evaluation and clinical validation of reconstructed DW images as well as computerized DT-MRI data such as tensor fields, eigen vector field, fiber tracks and parametric maps, we expect to achieve the first ultimate step in imaging in-vivo 3D fiber architecture of the human heart and speed up, for given MRI machine and equivalent image quality, the acquisition by a factor of at least 10 with respect to current SE-EPI scheme.