Hi everybody,
When running TBSS, are there any general requirements in terms of image
quality that should be considered? We have a long running study on infants
where there is a 15-dir (b=700 Voxel=1.4x1.4x2.2mm, TE/TR = 74/9000) diffusion
scan included, on which we would like to use TBSS.
A problem we keep on running into is that "tbss_2_reg -n" rotates our volumes
and parts of the brain are being cropped. Tracts seem to have lined up, but
the rotations are very unsatisfactory and alarming. In order to avoid these
problem we have tried pre-normalization and exclusion of individuals with
e.g. large ventricels, but the it does not seem to change anything.
Is there anyone else who has been experiencing these problems? or know what
they can originate from?
If someone would like to take a further look I can gladly send you some
exmaples (Attached png got reject due to size...).
|