High dynamic range fusion of magnetic resonance flow imaging data

Time-resolved, volumetric phase-contrast magnetic resonance imaging (also known as 4D flow MRI) is a comprehensive, diagnostic tool used by radiologists to simultaneously assess cardiovascular anatomy, flow, and function. Its ability to resolve both slow and fast hemodynamics (dynamic range) is limited, however. To address this, postprocessing techniques such as phase unwrapping have been used to extend the dynamic range, but are highly variable with respect to patient anatomy and physiology. Recently, a new technique called multi-venc 4D flow has gained traction due to its ability to simultaneously and robustly measure slow and fast flow. However, its use of the massive amount of data that it collects (10-20 GB) is inefficient. In this paper, I propose a high dynamic range (HDR) postprocessing technique for reconstructing and compressing multi-venc data down into digestible images that allow for more precise measurement of typical hemodynamic parameters measured from this data such as peak flow. I applied this technique to both simulated 4D flow MRI data, and invivo data. This technique shows promise for allowing comprehensive abdominal and neurovascular imaging, in which there is a large dynamic range of blood velocity.