Article Figures & Data
Figures
- Fig 1.
White matter EZ-tracing tractograms in a 24-year-old man. Images show the white matter tracts (yellow) superimposed on images obtained with b = 0 s/mm2.
A, Tractogram obtained by using conventional DTI.
B, Tractogram obtained by FLAIR DTI shows a larger area of white matter fiber tracts in both the genu and the splenium of the corpus callosum.
- Fig 2.
Anterior aspects of an image section obtained in an 23-year-old man show white matter tracts superimposed on images A–F magnified from the rectangular region-of-interest (solid rectangles). Partial volume effects near adjacent CSF lead to underestimation of FA in the genu of the corpus callosum on conventional DTI (left column and dashed rectangles in E.) Note the less consistent fiber directions and the darker gray level, which indicates lower FA values. These effects account for the smaller amount of fibers found with the tracking algorithm on conventional DTI than on FLAIR DTI (right column, dashed rectangles in F). Note the more consistent fiber directions and the brighter gray level, which indicates higher FA values.
A and B, Images obtained with b = 0 s/mm2 by using the EZ-tracing algorithm.
C and D, FA maps in gray scale displayed in the identical window level.
E and F, Vector-encoded FA maps. Red indicates left-right; green, anteroposterior; and blue, superoinferior.
- Fig 3.
Posterior aspects of the same image section as in Figure 2 show white matter tracts superimposed on the images in A–F magnified from the rectangular region-of-interest (solid rectangles). Although the FLAIR DTI tractogram shows more tracts in the splenium of the corpus callosum (B and dashed oval in F, with more consistent fiber directions) than the conventional DTI tractogram (A and dashed oval in E, with less consistent fiber direction near the lateral ventricle), false tracts are found in the occipital lobes (B).
A and B, Images obtained with b = 0 s/mm2 by using the EZ-tracing algorithm.
C and D, FA maps in gray scale displayed in identical window level obtained by using conventional (C) and FLAIR DTI (D). D is noisier than C, suggesting uncertainty in FA values.
E and F, Vector-encoded FA maps show that the false tracts in B are due to incidentally consistent orientations of the eigenvectors (rectangle in F), which is absent in E (rectangle in E). Red indicates left-right; green, anteroposterior; and blue, superoinferior.
- Fig 4.
Left anterior oblique views of white matter tracts (colored region) superimposed on the images in A and B. Images were obtained in another 23-year-old man by using the subvoxel tractography algorithm based on tensor deflection with identical parameter settings. Fiber tracts on FLAIR DTI were visually larger in volume than those on conventional DTI.
A, Original axial image (gray) obtained with seed points placed in the corpus callosum for conventional DTI.
B, FLAIR DTI.
- Fig 5.
Comparison of total volumes of fibers tracked on conventional DTI versus FLAIR DTI. For all subjects, FLAIR DTI depicted significantly more tracts than conventional DTI, with individual P values of .001–.02 (Student t test; n = 5 for each individual). On average, about 17% additional fiber tracts were detected on FLAIR DTI. The group difference was also significant (P < .0005, paired Student t test; n = 7).
