Unified reconstruction and motion estimation in cardiac perfusion MRI

We introduce a novel unifying approach to jointly estimate the motion and the dynamic images in first pass cardiac perfusion MR imaging. We formulate the recovery as an energy minimization scheme using a unified objective function that combines data consistency, spatial smoothness, motion and contrast dynamics penalties. We introduce a variable splitting strategy to simplify the objective function into multiple sub problems, which are solved using simple algorithms. These sub-problems are solved in an iterative manner using efficient continuation strategies. Preliminary validation using a numerical phantom and in-vivo perfusion data demonstrate the utility of the proposed scheme in recovering the perfusion images from considerably under-sampled data.