Abstract #3189

Ji J

Phase sensitive inversion-recovery (PSIR) has clinical applications ranging from myocardial infarction detection, brain tumor imaging, to fat/fluid suppression. Using phase-array coils, multi-channel IR data can be acquired and combined to improve the SNR (or to reduce scan time). However, conventional sums-of-squares reconstruction is not optimal for PSIR because it loses the phase information. This paper presents a tensor Markov random field model that can be used to reconstruct PSIR images from phased-array data. The new method does not require additional phase reference scans, or consistency correction of individual coil images. In-vivo brain MR experiments show that the method is robust and efficient.