Towards cerebrovascular 5D flow MRI

Motivation: Cardiac- and respiratory-phase resolved quantification of high-resolution 3D velocity vector fields for studying breathing-induced flow modulation in the brain and improving interventional treatment strategies. Goal(s): Implement and investigate feasibility of cerebrovascular 5D flow MRI Approach: Venous 5D flow MRI data were acquired during coached breathing, subsequently split into multiple configurations of two respiratory states, and reconstructed using state-of-the-art deep learning methods, followed by a time-domain analysis of various hemodynamic parameters. Results: Respiratory modulation of mean flow, amplitude, mean vorticity, and mean massic kinetic energy could be visualized and quantified for segments in healthy venous flow. Impact: We demonstrate the feasibility of high-resolution cerebrovascular 5D flow MRI of venous flow, facilitating quantification of respiratory modulation of 3D velocity fields and derived hemodynamic parameters, providing a framework for improved understanding of cerebral flow dynamics.