Yea, not sure. It could be that the uptake in WM near the NAcc is somewhat less than average (GTM assumes constant throughout). If so, then the GTM would subtract too much from the NAcc in an attempt to remove the spill in from WM. This is more of a problem for amyoild and tau ligands because then tend not to have clear anatomical delineations, esp in WM.
On 12/3/19 9:24 AM, miracle ozzoude wrote: > > External Email - Use Caution > > Thanks Doug. we double checked the registration with gtm.seg.mgz and > it looks good. Also the uptake in Acc is not much smaller than other > regions and is pretty similar on both hemispheres so we are not sure > what could be causing this. Any suggestions much appreciated. > Many thanks > > Paul > > On Mon, Dec 2, 2019 at 5:15 PM Greve, Douglas N.,Ph.D. > <dgr...@mgh.harvard.edu <mailto:dgr...@mgh.harvard.edu>> wrote: > > It may or may not be something that needs to be fixed. The PVC is > just a linear model where the regression coefficients are the > uptake in each ROI. As with any model, it adjusts the regression > coefs to minimize the error in the fit. If some end up being less > than zero when you really expect all to be > 0, then this probably > indicates an error in the (linear) model. This error could come > from several sources. For example, the registration could be off. > The PSF might not be right. Or the uptake might not be constant > across the ROIs. You should definitely start by checking the > registration. If the uptake in Nuc Acc is very small relative to > other ROIs, then I would just leave it the way it is, ie, DON'T > just set it to 0 when you go to a group analysis. > > On 12/2/2019 4:48 PM, miracle ozzoude wrote: >> >> External Email - Use Caution >> >> Hello Experts, >> >> I ran the PETsurfer pipeline using AV45 pet and performed pvc >> using pons as reference point. When i looked at the gtm.stats.dat >> file for one of my subjects, the SUVr for right and left >> nucleus-accumbens areas were in the negative (left = -0.361; >> right = -0.046). >> >> This is strange because SUVR shouldn't be negative. How do i go >> about fixing this error? Below, is my pvc command. >> mri_gtmpvc --i pet.nii --reg register.lta --psf 6 --seg >> gtmseg.mgz --default--seg-merge --auto-mask PSF .01 --mgx .01 --o >> subject_gtmpvc --rescale 174 >> >> Thanks. >> >> Best, >> Paul >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
