Using a state-of-the-art 1.5T MR system, the main goal of the proposed project is to create significant im- provements in spiral MRI via novel techniques that reduce imaging time, improve immunity to motion artifact, and improve off-resonance (and chemical shift) reconstruction so that new and/or improved cardiac, brain and body imaging applications will be enabled.
The specific aims are to: (1) improve spiral MR motion artifact immunity by developing efficient reductions in spiral MRI acquisition time through parallel imaging methods and reconstruction (e.g., SENSE, GRAPPA, conjugate gradient) at high acceleration; (2) develop new rapid off-resonance correction techniques with improved performance and reduced computation time when compared to conventional frequency segmented methods, (3) create robust spiral fat/water separation methods that prevent off-resonance blurring and provide superior performance and reduced acquisition time when compared to spatial spectral techniques, and, (4) develop methodologies for Pareto-optimal /(-space trajectory design based on minimizing time, aliasing energy, flow sensitivity, off-resonance blurring and/or other image quality measurements. Our project will show that use of new spiral acquisition, new post-processing methods and new trajectory de- sign techniques using/(-space and/or image data can reduce imaging time and significantly improve cardiac, brain and abdominal imaging. Preliminary results show that significant improvements in motion immunity through parallel imaging are possible. Significant work remains to explore improvements in reconstruction algorithms, amount of acceleration, and reliability. Off-resonance computational burden can be reduced by using a block re- gional correction method. For example, further improvements in performance and speed will explore use of re- gional information to control the computational block size. Our fat suppression methods will give rise to reductions in scan time by up to a factor of 2-4X with superior performance as compared to commonly used spatial spectral approaches; the efficiency/robustness of 2pt and 3pt Spiral Dixon methods (at equal scan time) are explored. These alone will enable improved abdominal and cardiac imaging when compared to conventional rectilinear or spiral methods. Uniquely, our advanced non-rectilinear trajectory, design method is that the inter-relationship be- tween off-resonance, motion and kappa-space sampling density effects are formally linked to image domain artifacts, thereby enabling unique waveform parameterizations and formal optimization to create superior trajectories (and hence better images) than those designed conventionally. We believe successful attainment of these aims will give rise to significant improvements in spiral MR and hence greater utilization of this important MR method.
|Yutzy, Stephen R; Seiberlich, Nicole; Duerk, Jeffrey L et al. (2011) Improvements in multislice parallel imaging using radial CAIPIRINHA. Magn Reson Med 65:1630-7|
|Bookwalter, Candice A; Griswold, Mark A; Duerk, Jeffrey L (2010) Multiple overlapping k-space junctions for investigating translating objects (MOJITO). IEEE Trans Med Imaging 29:339-49|
|Moriguchi, Hisamoto; Duerk, Jeffrey L (2006) Bunched phase encoding (BPE): a new fast data acquisition method in MRI. Magn Reson Med 55:633-48|