Hi,
If you just want a quick visualisation, then try dividing the original
image by the bias-corrected one (you'll need to set the options to
write it out, during the segmentation job), using imcalc. That could
have problems if there are any zeros in either image though, you might
want to set any non-finite values to 1...
If you need a more accurate estimate you'll need to load the
seg_sn.mat and try to reconstruct the bias field from the DCT
coefficients, using the code in spm_preproc and spm_preproc_write (the
transf subfunction in particular).
Good luck,
Ged
On 25/10/2007, <Maouia> <Bentlemsan> <[log in to unmask]> wrote:
> Hello,
> i've used spm5 for bias correction, but dont know how to visualize the bias
> field?
> Thanks in advance.
>
|