An inverse method to design RF coil arrays optimized for SENSE imaging

Abstract

A new method to design MRI RF coils that are optimized for SENSE (sensitivity encoding) imaging is introduced. In this approach, the inverse problem was solved where the surface current density distribution on a coil former was calculated to maximize the SNRsense within a volume of interest (VOI). For that purpose, an analytic relationship was formulated between the SNRsense and surface current density on the coil former. The SNR at pixel ρ in a SENSE-MR image, SNRsense,ρ, is inversely proportional to the g-factor: therefore, the g-factor was formulated in terms of the B1 distribution of the coils. Then, by specifying the geometry of the desired coil former and using a finite element mesh (FEM), the surface current distribution was calculated to maximize the SNRsense, by minimizing (1/SNRsense) in the VOI using a least squares procedure. A simple two-coil array was designed and built to test the method and phantom images were collected. The results show that the new coil design method yielded better uniformity and SNR in SENSE images compared to those of standard coils.