Functional Diffusion Tensor Imaging: Measuring Task-Related Fractional Anisotropy Changes in the Human Brain along White Matter Tracts (original) (raw)
Figure 5
Histogram distributions of voxels that are part of active fibers.
Here two histograms are shown of the number of ‘+’s per voxel, for all voxels along active fiber tracts, separately for positive correlation and negative correlation, combined for all subjects. In the case that there would not be an activity-related signal change, the distribution of +'s in these voxels would follow a normal distribution (solid red line). If only a small group of voxels (the active voxels) is responsible for a tract to be considered active then these voxels must show a high correlation (i.e. a high number of +'s) with the task while the distribution of +'s of the other (non-active) voxels that are part of the active tract would remain unchanged. This would result in a histogram built up from a large and a small normal distribution. The large group of non-active voxels would produce the original (large) normal distribution (solid red line) while the small group of active voxels would produce a second smaller distribution left (negative correlation) or right (positive correlation) from the original distribution. However, if not a small group but the majority of the voxels along the active fibers belong to the set of active voxels, one would expect a single normal distribution which is shifted to left (negative correlation) or to the right (positive correlation). Left, the histogram for the (positive) active voxels for the tactile task is shown while on the right side the histogram for the (negative) active voxels of the visual task is shown. Both histograms appear to follow a single normal distribution but are shifted to the right (positive correlation with tactile task) or to the left (negative correlation with visual task). This suggests that the task-related signal changes found with fDTI indeed occur along large parts of the fibers and are not confined to small parts of the fibers.