Premium
Accelerating parameter mapping with a locally low rank constraint
Author(s) -
Zhang Tao,
Pauly John M.,
Levesque Ives R.
Publication year - 2015
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.25161
Subject(s) - rank (graph theory) , constraint (computer aided design) , flip angle , computer science , variable (mathematics) , algorithm , acceleration , artificial intelligence , mathematics , pattern recognition (psychology) , magnetic resonance imaging , physics , mathematical analysis , geometry , combinatorics , medicine , classical mechanics , radiology
Purpose To accelerate MR parameter mapping using a locally low rank (LLR) constraint, and the combination of parallel imaging and the LLR constraint. Theory and Methods An LLR method is developed for MR parameter mapping and compared with a globally low rank method in a multiecho spin‐echo T 2 mapping experiment. For acquisition with coil arrays, a combined LLR and parallel imaging method is proposed. The proposed method is evaluated in a variable flip angle T 1 mapping experiment and compared with the LLR method and parallel imaging alone. Results In the multiecho spin‐echo T 2 mapping experiment, the LLR method is more accurate than the globally low rank method for acceleration factors 2 and 3, especially for tissues with high T 2 values. Variable flip angle T 1 mapping is achieved by acquiring datasets with 10 flip angles, each dataset accelerated by a factor of 6, and reconstructed by the proposed method with a small normalized root mean square error of 0.025. Conclusions The LLR method is likely superior to the globally low rank method for MR parameter mapping. The proposed combined LLR and parallel imaging method has better performance than the two methods alone, especially with highly accelerated acquisition. Magn Reson Med 73:655–661, 2015. © 2014 Wiley Periodicals, Inc.