Comprehensive approach for correction of motion and distortion in diffusion-weighted mri. Magnetic Resonance in Medicine

We present a method for removing artifacts related to patient motion and eddy-current induced image distortions in DW-MRI. The images are corrected for motion and distortion simultaneously using a well-established registration technique. Subject motion is modeled using a 3D rigid body transformation while image distortion is modeled using a general solution for eddy-current induced fields inside the magnet. The DW-MRI dataset can at the same time be rigidly aligned to a standard template. Thus our robust post-processing method can be used to simultaneously remove motion and distortion related artifacts as well as to position the dataset in a standardized orientation. Image misalignment can be a significant source of artifacts in data analysis and it reduces the clinical and biological utility of the dataset. Though post-acquisition methods to remove image distortion have existed for some time [1], until recently, the problem of patient motion during DW-MRI acquisitions was commonly ignored [2]. The method we propose achieves two objectives simultaneously and automatically: motion and distortion correction of the DW-MRI dataset as well as the positioning of the dataset i