Hi,
The problem is only in those two slices. I created a very simple mask (w4)
from one "trouble-slice" by giving value of 10 to the posterior part of the
brain, one elsewhere, and zeros to the "nonbrain".
I called flirt like:
/opt/fsl/bin/flirt -ref Jaak4 -refweight w4 -in epi4 -out epiw4 -2D
-schedule /opt/fsl/bin/sch_2D_5dof -paddingsize 1
But the result was basically the same. Images are as good as the upper
slices, so that shouldn't be the problem.
Regards,
Pasi
|