Needless to say, MJ is right indeed!
Potentially problematic negative values coming from a
negative Jacobian determinant will disappear in the
future version of FSLVBM, as we will use the new
non-linear registration FNIRT...
Cheers,
Gwenaëlle
--- "Zhang, Xiaochu (NIH/NIDA) [F]"
<[log in to unmask]> a écrit :
> Thank you so much, Mark!
> I just checked my data. It is same to your
> assumption that most of my value is positive and
> grey matter part is usually around 1000 (most are
> lower than 1000, about several hundreds).
> Now, I feel much better to my data.
> Thanks, again!
> Xiaochu
> -----Original Message-----
> From: Mark Jenkinson [mailto:[log in to unmask]]
> Sent: 2008ǯ1·î2Æü 12:23
> To: [log in to unmask]
> Subject: Re: [FSL] The unit of VBM result
>
> Hi,
>
> Gwen or others can correct me if I'm wrong, but I
> think that all
> segmentation results
> in fslvbm are scaled by a factor of 1000 in the
> native space and then
> for the _mod_
> images they are again scaled by the local Jacobian.
>
> If this is the case then values up to 5233 are
> possible and not
> unreasonable,
> although I'd expect most of your voxels to have
> values more like 1000
> in grey matter.
> Negative values are worse and probably indicate
> negative Jacobian
> values which
> can occur but should be very rare and hopefully in
> uninteresting
> places. Check
> that you do not have many voxels with values less
> than zero.
>
> If most of your values are positive and of the order
> of 1000 in grey
> matter, then
> I think everything is probably fine.
>
> All the best,
> Mark
>
>
>
> On 1 Jan 2008, at 20:17, Zhang, Xiaochu (NIH/NIDA)
> [F] wrote:
>
> > Thank you very much for your response!
> >
> > However, I just use fslstats to check my original
> data (i.e.,
> > ¡ÈGM_mod_merg_s3.5.nii.gz¡É). It is form -462 to
> 5233!
> > Is there something wrong?
> >
> > Could you please give me some suggestions which
> one I need check more?
> >
> > Thank you very much!
> >
> > Xiaochu
> >
> > From: Lynne Gauthier [mailto:[log in to unmask]]
> > Sent: 2007ǯ12·î31Æü 14:05
> > To: [log in to unmask]
> > Subject: Re: [FSL] The unit of VBM result
> >
> >
> >
> > The value of each voxel from the segmentation
> should be between 0
> > and 1. Closer to 1 means more grey matter.
> Perhaps you have the
> > sum for all the voxels in the ROI, not the
> average?
> >
> >
> >
> > On Dec 30, 2007, at 4:37 PM, Zhang, Xiaochu
> (NIH/NIDA) [F] wrote:
> >
> >
> >
> > Hi FSL experts,
> >
> > I am working on the VBM analysis on my data. Now I
> want to do some
> > individual analysis. So I got the mean value of
> one ROI in each
> > subject. The value is usually about several
> hundred. I think the
> > higher value means the higher density of grey
> matter. Is it correct?
> >
> > Additionally, I was confused on the unit of these
> values. I read
> > some VBM papers and found their data is usually
> 0-1.
> >
> > who can help me?
> >
> > Thank you very much
> >
> > Xiaochu
> >
> >
> >
> > Lynne Gauthier
> >
> > Graduate Research Associate
> >
> > Neuroimaging Lab
> >
> > CI Therapy Research Group
> >
> > University of Alabama - Birmingham
> >
> > 857-998-7785
> >
> > [log in to unmask]
> >
> >
> >
> >
> >
> >
> >
> >
> >
>
_____________________________________________________________________________
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail http://mail.yahoo.fr
|