Abstract
In diffusion-weighted magnetic resonance imaging (dMRI), the coordinate systems where the image and the diffusion gradients are represented may be inconsistent, thus impacting the quality of subsequent fiber tracking and connectivity analysis. Empirical verification of the reconstructed fiber orientations and subsequent correction of the gradient table (by permutation and flipping), both time-consuming tasks, are therefore often necessary. To save manual labor in studies involving dMRI, we introduce a new automatic gradient-table verification approach, which we propose to include in the dMRI processing pipeline. To that end, we exploit the concept of fiber continuity – the assumption that, in the fibrous tissue (such as the brain white matter), fiber bundles vary smoothly along their own orientations. Our tractography-free method tries all possible permutation and flip configurations of the gradient table, and in each case, assesses the consistency of the reconstructed fiber orientations with fiber continuity. Our algorithm then suggests the correct gradient table by choosing the configuration with the most consistent fiber orientations. We validated our method in 185 experiments on human brain dMRI data form three public data sources. The proposed algorithm identified the correct permutation and flip configuration for the gradient table in all the experiments.
Highlights
An automated method has been proposed[3] to align the coordinate frame of the gradient orientations with that of the corresponding diffusion weighted images, which uses the average fiber trajectory length computed from whole brain fiber www.nature.com/scientificreports/
We implemented our method in Matlab, and validated it in 185 experiments by analyzing existing de-identified human brain diffusion-weighted magnetic resonance imaging (dMRI) data from three public datasets, each containing images acquired with a number of different b-values (Table 1)
As opposed to the dwigradcheck command implemented in MRtrix[33], our public implementation is in Matlab, making it a good candidate for automatic inspection of dMRI data that is loaded for further analysis in Matlab
Summary
An automated method has been proposed[3] to align the coordinate frame of the gradient orientations with that of the corresponding diffusion weighted images (considering flips, permutation, and small rotations due to angulation of the acquisition plane), which uses the average fiber trajectory length computed from whole brain fiber www.nature.com/scientificreports/. Another similar automatic tool[8] suggests the appropriate flipping (though not the permutation) of the gradient table by counting the number of long tracts in the tractography results produced with each possible flip configuration. To ensure that repeated tractography in various alignment configurations does not result in a long runtime, algorithmic optimizations can be used in the whole-brain tractography routine[3], or local tractography can be performed in a region of interest using simpler diffusion models. We introduce a new automatic algorithm that, without using tractography, recommends the necessary permutation and flipping of the gradient table from the dMRI signal. We compute the fiber continuity error with all possible permutation and flip configurations for the gradient table, and choose the configuration that produces the least error. We evaluate the performance of our method on human brain dMRI data from three public data sources
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
