We find these artifacts in our DTI images (most visible in the b0 images as
shown in the attachment). For lack of a better term, we are calling them
modulation artifacts; they go left-right in the frequency encode direction and
appear as light/dark alternating voxel by voxel.
They do not show up on a phantom as far as we can tell. So we're left
wondering if it is an artifact due to blood flow or pulsatility? It seems they
are most prominent in the center of the brain, near ventricles and
subcortical structures. We are using an eddy-current compensated double
spin-echo, echo-planar pulse sequence adapted from Reese et al. (2003).
Beyond wondering what the source is, we also wonder if this is a cause for
concern (does it degrade our DTI data in those regions), and if there is a fix
for this.
Here's a link to the image in case the attachment does not work:
http://www.princeton.edu/~mpinsk/modulation_artifact.jpg
Thanks! (sorry if this is posted twice accidentally)
Mark
|