A sparse bayesian learning for highly accelerated dynamic MRI

Abstract

In dynamic MRI, spatio- temporal resolution is a very important issue. Recently, compressed sensing approach has become a highly attracted imaging technique since it enables accelerated acquistion without aliasing artifacts. Our group has proposed an l(1)-norm based compressed sensing dynamic MRI called k-t FOCUSS, which outperforms existing methods. However, it is known that the restrictive conditions for l(1) exact reconstruction usually cost more measurements than l(0) minimization. In this paper, we adopts a sparse Bayesian learning approach to improve k-t FOCUSS and achieve l(0) solution. We demonstrated the improved image quality using in vivo cardiac cine imaging.