This research project ?Radial Echo Volumar Imaging? proposes the development of MRI acquisition and reconstruction methods based on a novel versatile non-Cartesian sampling concept for fast motion-corrected imaging. The technique expands upon Echo Planar Imaging (EPI), which is the most widely utilized fast MRI acquisition and is the standard method for various applications ranging from functional MRI to diffusion and perfusion imaging. Recently EPI has also been shown to be promising for rapid structural imaging including simultaneous multi-parametric MRI. Most modern EPI approaches are based on volumetric imaging methods as they permit high isotropic spatial resolution, improved Signal to Noise Ratio per unit time, and parallel imaging acceleration along the third dimension. A challenge of volumetric imaging however is the requirement for segmentation due to gradient and physiological limitations that leads to increased motion sensitivity and other physiological effects. Radial sampling offers several advantages with regards to segmented acquisitions including robustness to motion due to intrinsic self-navigation from oversampling the center of k-space. Radial sampling also has the benefit of producing benign ?streaking? aliasing artifacts compared to Cartesian allowing for large accelerations and an efficient use of parallel imaging methods. A further advantage of radial sampling is that the in-plane dimension is sampled quickly by frequency encoding leading to higher in-plane resolution and less distortion with low time penalty. In this project we propose to utilize these advantages to develop innovative methodology for rapid and robust brain imaging that should also prove to be important for many other imaging applications including body MRI. The Scientific Premise of this proposal is that an optimal rapid MRI acquisition can be obtained by using three- dimensional radial EPI trajectories and generalized model-based reconstructions. We propose an innovative Radial Echo Volumar Imaging (REVI) acquisition created by adding gradient encoding along the third direction of a radial EPI acquisition to create SMS and 3D rotated ?Stack-of-Stars? sampling for high parallel imaging acceleration while allowing for optimal tradeoffs in temporal and spatial resolutions. The self-navigation properties of the radial trajectories will provide motion robustness and continuous golden-angle rotation will permit variable temporal resolutions and reordering of the acquisition. The multi-echo nature of REVI will also allow for simultaneous multi-parametric structural scanning. The proposed technology requires no special hardware and can be run on any scanner by any investigator.
This is a proposal for the development of pulse sequence and image reconstruction software for fast and motion robust magnetic resonance imaging (MRI). Speed and motion artifacts are limiting factors in MRI and new methods that take advantage of the latest technology will have a significant impact on healthcare. In particular we are proposing novel ?Radial Echo Volumar Imaging? methods for brain MRI that are capable of high spatial resolutions with reduced artifacts related to motion in short acquisition times. The methodology is also directly applicable numerous other applications including free breathing body imaging. Any researcher can run the proposed developments on any MRI scanner without the need for extra hardware or expense.
