Premium
Motion‐corrected k‐space reconstruction for interleaved EPI diffusion imaging
Author(s) -
Dong Zijing,
Wang Fuyixue,
Ma Xiaodong,
Dai Erpeng,
Zhang Zhe,
Guo Hua
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.26861
Subject(s) - diffusion mri , computer science , ghosting , rotation (mathematics) , translation (biology) , computer vision , artificial intelligence , diffusion , fractional anisotropy , magnetic resonance imaging , orientation (vector space) , echo planar imaging , motion (physics) , nuclear magnetic resonance , physics , mathematics , geometry , chemistry , medicine , biochemistry , messenger rna , gene , radiology , thermodynamics
Purpose To develop a new approach to correct for physiological and macroscopic motion in multishot, interleaved echo‐planar diffusion imaging. Theory This work built on the previous SPIRiT (iterative self‐consistent parallel imaging reconstruction) based reconstruction for physiological motion correction in multishot diffusion‐weighted imaging to account for macroscopic motion. In‐plane rotation, translation correction, data rejection, and weighted combination are integrated in SPIRiT‐based reconstruction to correct for ghosting artifacts, blurring, altered b‐matrix, and residual artifacts caused by motion. Methods Numerical simulations (one data set was obtained from the Human Connectome Project) and in vivo experiments with deliberate bulk motion were performed to demonstrate the effectiveness of the proposed method. Diffusion images and quantitative tensor parameters were calculated to evaluate the correction performance. Results The proposed method provided images with reduced artifacts and diffusion tensors with improved accuracy in both simulations and in vivo experiments. For in vivo experiments with deliberate motion, the percentage error of fractional anisotropy in the genu of the corpus callosum was significantly reduced from 17.01 ± 12.64 to 5.73 ± 3.77 through motion correction. Conclusions The proposed method can effectively correct for physiological and macroscopic motion artifacts in multishot interleaved echo‐planar imaging, generate high resolution diffusion images, and improve the accuracy of tensor calculation. Magn Reson Med 79:1992–2002, 2018. © 2017 International Society for Magnetic Resonance in Medicine.