Indeed, no obvious problem here. Could you please tgz and upload your stats/ directory? Also, could you specify which version of FSL you are using?
Thanks,
Gwenaelle
--- En date de : Jeu 28.5.09, Michael Okujava <[log in to unmask]> a écrit :
> De: Michael Okujava <[log in to unmask]>
> Objet: Re: [FSL] Re : [FSL] Problem with FSLVBM statistics
> À: [log in to unmask]
> Date: Jeudi 28 Mai 2009, 15h01
> Hi Gwenaelle
>
> thank you for the suggestion. You will find the design
> matrix and contrast
> files attached. there are 18 cases, first nine are
> controls.
>
> Thanks
>
> Michael
>
>
>
>
> On Thu, 28 May 2009 05:14:17 -0700, Gwenaëlle DOUAUD
> <[log in to unmask]>
> wrote:
>
> >Hi,
>
> I don't think bet can be related in any way to your absence
> of corrp maps...
> Could you please send your design.mat and design.con? Also,
> which version of
> FSL are you using exactly?
>
> Cheers,
> Gwenaelle
>
> --- En date de : Jeu 28.5.09, Michael Okujava <[log in to unmask]>
> a écrit :
>
> > De: Michael Okujava <[log in to unmask]>
> > Objet: Re: [FSL] Re : [FSL] Problem with FSLVBM
> statistics
> > À: [log in to unmask]
> > Date: Jeudi 28 Mai 2009, 13h42
> > Hi
> >
> > the tstats maps look corrupt. and the raw tstat maps
> -
> > mod_merg_s* files -
> > are also too "noisy". But there is no error message
> during
> > the processing.
> > I would like to doublechek if the insufficient brain
> > extraction is causing the
> > errors (even though I can't understand why it should
> change
> > the format of
> > output of statistics map (not giving tfce_corrp_tstat
> > files). Which additional
> > BET command will you recommend to use with -N to
> extract
> > the brain more
> > precisely (in recent messages -R function was
> mentioned not
> > changing a lot)?
> >
> > Thank you
> >
> > Michael
> >
>
>
>
>
>
|