Obtaining B1 Distributions by Encoding in B1 Instead of Image Space

PURPOSE: The BEAR method is a recently proposed phase-based B1 mapping method. An important feature of BEAR is that the phase sensitivity to B1 can be manipulated with sequence parameters. We describe a new method that calculates the B1 distribution in a volume from multiple 1D BEAR projections, where each projection has a different B1 sensitivity. In essence, we are replacing spatial phase encoding with B1 phase encoding. An estimate of the B1 distribution in each projected pixel is then calculated using a convex optimization formulation. With this method, an estimate of the B1 distribution in a volume can be attained faster than by acquiring a 2D B1 map. Additionally, this method may resolve rapid B1 variations in space better than with a 2D image. We validate this method through simulations and in vivo at 3T.