Meeting Banner
Abstract #3484

Feasibility of Accelerating 3 T Hip Imaging Using Compressed Sensing

Riccardo Lattanzi1, 2, Alicia W. Yang1, 2, Li Feng, 23, Michael P. Recht4, Daniel K. Sodickson1, 2, Ricardo Otazo1, 2

1Radiology/Center for Biomedical Imaging, NYU Langone Medical Center, New York, NY, United States; 2The Sackler Institute of Graduate Biomedical Sciences, New York University School of Medicine, New York, NY, United States; 3Radiology/Center for Biomedical Imaging, New York University Langone Medical Center, New York, NY, United States; 4Radiology, NYU Langone Medical Center, New York, NY, United States

Magnetic resonance imaging (MRI) of the hip joint requires high-spatial resolution and signal-to-noise ratio (SNR), which results in long scan times. Compressed sensing (CS) reconstruction approaches are currently used in research to achieve very large accelerations for advanced MR studies. We explored the feasibility of using CS to achieve moderate acceleration factors in routine hip imaging, while maintaining diagnostic quality. Our results suggest that it could be possible to accelerate 2D turbo spin echo acquisitions by a factor of 4 using CS and that 6-fold acceleration is also feasible for T2 mapping based on multi spin echo time-series.

Abstract PDF Presentation Video

Keywords

accelerate accelerated acceleration accelerations acceptable achieve acquisition acquisitions additional allow almost applied approaches array articular axial bandwidth better biochemical biomedical cartilage central challenging chose chosen clinical clinically coil coils column combination component compressed coronal critical currently datasets decay decreasing degree density diagnostic dimensional distribution echoes evaluate exam exploit exponential feasibility feasible female fitting flexible fold folds full fully function furthermore future graduate healthy identical implementing improving includes incoherence institution intensity iterative joint joints least lengthen long maintained maps materials matrix medical medicine moderate near nine noise noisier nonlinear normally orthogonal outer oversampling parallel pattern pelvis pixel principal protocol qualitatively quality radiology random receive reconstructed reconstructing reconstruction reduce representative required requires resolution retrospective routine sampled sampling saturation saved scanned school sciences section selected sensing series several side slice slices solutions space sparsity spatial spin spine square studies suggests superimposed system temporal thresholding torso transform transmission truly turbo variable various white wrapping yang