Title
Comparison of EPI distortion correction methods in diffusion tensor MRI using a novel framework.
Abstract
Diffusion weighted images (DWIs) are commonly acquired with Echo-planar imaging (EPI). B0 inhomogeneities affect EPI by producing spatially nonlinear image distortions. Several strategies have been proposed to correct EPI distortions including B0 field mapping (B0M) and image registration. In this study, an experimental framework is proposed to evaluation the performance of different EPI distortion correction methods in improving DT-derived quantities. A deformable registration based method with mutual information metric and cubic B-spline modeled constrained deformation field (BSP) is proposed as an alternative when B0 mapping data are not available. BSP method is qualitatively and quantitatively compared to B0M method using the framework. Both methods can successful reduce EPI distortions and significantly improve the quality of DT-derived quantities. Overall, B0M was clearly superior in infratentorial regions including brainstem and cerebellum, as well as in the ventral areas of the temporal lobes while BSP was better in all rostral brain regions.
Year
DOI
Venue
2008
10.1007/978-3-540-85990-1_39
MICCAI (2)
Keywords
Field
DocType
experimental framework,deformable registration,dt-derived quantity,epi distortion,diffusion tensor mri,b0m method,diffusion weighted image,epi distortion correction methods,different epi distortion correction,image registration,novel framework,spatially nonlinear image distortion,bsp method,mutual information
Computer vision,Diffusion MRI,Distortion correction,Nonlinear system,Pattern recognition,Echo-planar imaging,Data mapping,Computer science,Mutual information,Artificial intelligence,Image registration
Conference
Volume
Issue
ISSN
11
Pt 2
0302-9743
Citations 
PageRank 
References 
26
1.31
2
Authors
7
Name
Order
Citations
PageRank
Minjie Wu1745.16
Lin-Ching Chang28312.36
L. Walker31769.21
H Lemaitre4322.47
Alan S. Barnett5543.79
Stefano Marenco6512.70
Carlo Pierpaoli746748.78