The goal of this work is to investigate novel acquisition and reconstruction strategies to significantly increase the performance of contrast-enhanced magnetic resonance angiography (CE-MRA). Though CE- MRA is less invasive than X-ray angiography (XRA) and utilizes no ionizing radiation, XRA has considerably higher spatial and temporal resolution.
Our aim i s to develop, optimize, and evaluate new methods using projection reconstruction (PR) techniques that deliver higher spatial resolution per unit time than conventional Fourier k-space traversals. Efficiency will be increased by limiting the number of projection angles acquired. Sparse sampling in X-ray computed tomography (CT) using filtered back projection (FBP) produces """"""""clutter"""""""" artifacts which are large relative to the small differences in tissue attenuation displayed by CT. Preliminary results suggest that the large signal difference between enhanced vessels and static tissue in CE-MRA make these PR artifacts acceptable in some applications. Since the vessels generate the clutter artifacts, estimates of the artifacts can be generated and later reduced in post-processing. This PR strategy provides a preliminary four-fold reduction in scan time. Coupled with k-space scheduling and reconstruction methods for high temporal resolution, high spatial resolution can be provided at several key time frames throughout the arterial contrast enhancement curve. Three main 3DPR acquisition techniques will be developed to provide flexibility when prescribing spatial and temporal resolution. The proposed work includes simulations and phantom experiments to optimize the acquisition strategies and quantify their performance. Algorithms for artifact reduction will be further investigated and refined during these experiments as well. The optimized techniques will be clinically evaluated in the carotid arteries, lungs and abdomen. Successful completion of this work will place CE-MRA as a stronger competitor to the more invasive X-ray DSA technique.
