Thanks for the reply. Just to make sure I got the right files. After
processing with FEAT I get several images in the subject folders. Are
the right contrast images called for instance:
thresh_zstat1.nii.gz
if it is the 1st individual contrast I want to use in the group analysis?
All the best,
Julian
On Fri, Sep 16, 2016 at 10:59 PM, Mark Jenkinson
<[log in to unmask]> wrote:
> Hi,
>
>
>
> Yes, for group-level analysis you just assemble the COPE images from the
> first level, but after being transformed to standard space, into a single 4D
> file which is then the input to randomise.
>
>
>
> All the best,
>
> Mark
>
>
>
> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
> Julian Macoveanu <[log in to unmask]>
> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
> Date: Friday, 16 September 2016 at 09:57
> To: "[log in to unmask]" <[log in to unmask]>
> Subject: Re: [FSL] Randomise in fMRI
>
>
>
> Thanks for the reply. Just to be clear, since I did not find any
>
> examples on using randomise on fMRI data... for a group level analysis
>
> then, I simply construct a 4D file made from the individual 1st level
>
> contrasts?
>
>
>
> Best,
>
> Julian
>
>
>
> On Fri, Sep 16, 2016 at 10:17 AM, Anderson M. Winkler
>
> <[log in to unmask]> wrote:
>
> Hi Julian,
>
>
>
> Because the temporal autocorrelation in FMRI timeserie, randomise isn't
>
> supposed to be used with 1st level analyses, only group level.
>
>
>
> All the best,
>
>
>
> Anderson
>
>
>
>
>
> On 15 September 2016 at 09:05, Julian Macoveanu <[log in to unmask]>
>
> wrote:
>
>
>
> Dear all,
>
>
>
> I am considering non-parametric estimation of fMRI data using
>
> randomise. I am however not sure how to construct the 4D input file. I
>
> have made all individual 4D files which I used for the FEAT analysis.
>
> Also not sure how to inform randomise on the onset times.
>
>
>
> Thankful for any input,
>
> Julian
>
>
>
>
>
>
|