Abstract
View-sharing techniques have been proposed to increase the frame rate for dynamic imaging. View-sharing PROPELLER (VS-Prop) updates the central k-space continuously by using rotating blades, whereas the outer k-space data are provided by adjacent blades acquired at lower frame rates to achieve high effective spatial resolution. Image reconstruction of fine structures contributed more from the high-spatial-frequency components is thus sensitive to the priority decision or neighboring assignment [3] of blades. Since the dynamic signal response can be spatially varying, manipulation of the composing blades in VS-Prop does not have to be identical for different regions. In this study, therefore, we propose a novel method termed pixel-based optimal blade selection (POBS) to minimize reconstruction artifacts due to under-sampled outer k-space data.