Inherent insensitivity to B1 field inhomogeneity using regularized nonlinear inversion reconstruction

INTRODUCTION Signal intensity variation that arises from inhomogeneous transmit and receive fields not only complicates the qualitative appearance of MR images but also presents the major source of error in quantitative MRI/S. Many current approaches address this problem by mapping the transmit field [1] and assuming the same transmit and receive field patterns during compensation [2]. However, this assumption becomes problematic at high field where the correction becomes more significant [3]. While work has also been done to simulate both fields numerically, these calculations are usually model-dependent and cannot account for field inhomogeneity due to non-RF field factors. Collins et al. observed that with known receptivity, some parallel imaging methods such as SENSE can remove the receptivity inhomogeneity [4]. However, the coil receptivity is generally unknown and needs to be estimated from additional acquisitions. Recently, Ying [5] and Uecker [6] each proposed a method to jointly reconstruct the images and estimate the coil sensitivity for parallel imaging. This work studies the effect of the latter regularized nonlinear inversion approach on image signal intensity. We demonstrate that not only receptivity inhomogeneity, but also transmit inhomogeneity may be removed simultaneously. Therefore this approach may play a vital role in both quantitative and qualitative analysis with MRI beyond its application for parallel imaging.