High-resolution imaging using partial separability of spatiotemporal signals with a novel data sampling scheme

Guoxi Xie,Xiang Feng,Xin Liu,Bensheng Qiu,Anthony G. Christodoulou

2011-01-01

Abstract:Introduction: The Partial Separability (PS) of spatiotemporal signals has been exploited for sparse sampling of (k, t)-space to speed up MRI [1,2]. PS model-based sparse sampling schemes often collect two datasets, one (called image data) with extended k-space coverage to provide the desired spatial resolution and the other (called navigator data) with high temporal resolution to define the temporal subspace. Based on the theory of partially separable functions, the more spatial frequency components in (k-f) space the navigator data covers, the more accurately the PS model will capture spatiotemporal changes. To address this issue, we present a novel sampling method that uses radial sampling trajectories for collecting the navigator data and Cartesian sampling trajectories for collecting image data. Simulation and in vivo cardiac imaging results demonstrate that the proposed method could produce much better reconstructions. Theory: Based on the theory of partially separable functions [3], if a signal ,t) k S( v is dependent on both space r and time t , it can be decomposed as