You can also use SPM8-New Segment to skull strip. Like most skull
stripping methods, there's a degree of individual level fine tuning
required. I usually use something like:
- Sum c1 (grey matter) and c2 (white matter) into a single image and
binarise at 0.1 (i.e. in imcalc "(i1+i2)>0.1")
- Smooth by 1-2mm
- Re-binarise at 0.2 (i1>0.2)
- Multiply the mask by the image you wish to strip (i1.*i2)
Whether or not you use the CSF (c3), though sometimes it can be a bit
overinclusive in terms of eyeballs.
Obviously you need to do this with the anatomical image, and then
register everything to your functional.
On 2/14/12, Suehyun Kim <[log in to unmask]> wrote:
> Hi, SPM users
> I have functional and anatomical imaging data.
> I want to analyse normalisation with skull stripping.
> Can I use BET2 to extract skull using spm8? and How?
> I searched here mail list and found some answers.
> someone attached a BET algorithm(mat file). so, i tested this.
> but it was for spm5 and had to use cygwin(do not support for FSL anymore)
> instead of vmware player.
> i'd like to use SPM8 for preprocessing but also want to use BET for skull
> How can I do that?
> Many thanks
> S, Kim