> I have a quick question regarding image orientation. I am feeding smoothed
> normalized realigned files into first-level statistical analysis. In these
> images, I am sure that RIGHT is RIGHT. If I run these images through 1st
> level model estimation and get the appropriate contrast files, are these
> going to be in the same orientation, i.e. RIGHT is RIGHT? I do not know if
> this matters but the
> spm_default file says: defaults.analyze.flip = 0;
Do you mean that the images are displayed using Display or Check Reg with the
right side of the brain on the right side of the axial image? If so, then
you should be able to run these data through any of the SPM utilities and
still have the right side of the brain on the right side of the image. That
is unless you rotate by 180 degrees, in which the left and right sides may be
swapped (as will one other direction). A flip (which can be done via the
Display button if either one or three negative zooms are applied) will also
reverse the sides, but you would be warned about this. I can't say what
would happen if you used stuff outside SPM though.
Best regards,
-John
|