Premium
Coherence enhancement in quantitative susceptibility mapping by means of anisotropic weighting in morphology enabled dipole inversion
Author(s) -
Kee Youngwook,
Cho Junghun,
Deh Kofi,
Liu Zhe,
Spincemaille Pascal,
Wang Yi
Publication year - 2018
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.26748
Subject(s) - weighting , isotropy , quantitative susceptibility mapping , anisotropy , orientation (vector space) , nuclear magnetic resonance , mathematics , algorithm , physics , artificial intelligence , magnetic resonance imaging , computer science , optics , geometry , acoustics , medicine , radiology
Purpose To investigate an anisotropic structural prior in morphology enabled dipole inversion (MEDI) for improving accuracy in quantitative susceptibility mapping (QSM). Theory and Methods Anisotropic weighting (AW) was devised and implemented to incorporate orientation information into the edge agreement in the MEDI method. AW performance was compared with isotropic weighting by testing and validating on in vivo brain multiple orientation MRI data using COSMOS and the (33) component of the susceptibility tensor as reference. Results Suppressing streaking artifacts, AW improved not only QSM image quality but also accuracy in terms of RMSE (root mean square error), HFEN (high frequency error norm), SSIM (structural similarity index), and GDA (gradient direction agreement). In addition, it outperformed isotropic weighting in region of interest‐based analysis. From a computational perspective, AW was as fast as isotropic weighting, taking approximately the same central processing unit times. Conclusion Using AW in MEDI improves QSM accuracy compared with isotropic weighting. Magn Reson Med 79:1172–1180, 2018. © 2017 International Society for Magnetic Resonance in Medicine.