Non-uniformity correction of human brain imaging at high field by RF field mapping of B1+ and B1-

A new method of non-uniform image correction is proposed. Image non-uniformity is originated from the spatial distribution of RF transmission and reception fields, represented as B(1)(+) and B(1)(-), respectively. In our method, B(1)(+) mapping was performed invivo by a phase method. In B(1)(-) mapping, images with multiple TEs were acquired with a multi-echo adiabatic spin echo (MASE) sequence which enables homogeneous excitation. By T(2) fitting of these images an M(0) map (M(0)(MASE)) was obtained, in which signal intensity was expressed as the product of B(1)(-) and M?(1-e?(TR/T1)) . The ratio of this M(0)(MASE) map to the B(1)(+) map showed a similar spatial pattern in different human brains. These ratios of M(0)(MASE) to B(1)(+) in 24 subjects were averaged and then fitted with a spatially polynomial function to obtain a ratio map of B(1)(-)/B(1)(+)(α). Uniform image was achieved in spin echo (SE), MASE and inversion recovery turboFLASH (IRTF) images using measured B(1)(+) and calculated B(1)(-) by αB(1)(+). Water fractions in gray and white matters obtained from the M(0) images corrected by this method were in good agreement with previously reported values. From these experimental results, the proposed method of non-uniformity correction is validated at 4.7 T imaging.