Dear Francisco,
> In preprocessing of structural data (VBM), you should do Bias correction, right?, but when you do the segmentation, I think that SPM do another bias correction. Are both bias correction necesaries?, if you don't do the first bias correction, are the data wrong?
Inhomogeneities (bias) in the signal will affect segmentation, and so
need to be corrected for. You are correct that bias-correction is
included in the segmentation routine. However, in some cases, this
doesn't fully remove the bias. In these cases an easy solution is to
run bias-correction twice (as is often suggested on the SPM list).
So, if segmentation is working, then there is no need to run a
two-pass bias correction. There is nothing wrong with this. If,
however, you find that segmentation is failing, and you have also
tried repositioning your structural scans to provide good starting
estimates for spatial registration (also frequently recommended), then
the two-pass bias correction method may help.
Hope this helps!
Best regards,
Jonathan
--
Jonathan Peelle, PhD
Department of Otolaryngology
Washington University in St. Louis
Office: (314) 362-9044
http://peellelab.org || http://jonathanpeelle.net
|