 |
 |
Hi - I'm not sure which images AFNI uses for the actual inference, but FEAT uses Z not t (it converts from t to Z before doing thresholding). This probably explains the differences you are seeing. Also, note that cluster-based thresholding may reject some clusters of supra-Z-threshold voxels if they are not big enough (where "big" is determined by Gaussian random field theory). Cheers, Steve. On Tue, 24 Aug 2004, Jacqueline Horwitz wrote: > Hello again, > I do have the tstat image selected for thresholding. The activations appear > to be centered around the same general areas as afni shows, however they are > much more expansive, and for some reason survive when I type in a ridiculous > threshold of 20. Considering that the activations barely survive a > threshold of 2 in afni, and they did not survive the clustered thresholding > in the FEAT report at all, I don't quite know what could be causing that? > J > > On 8/23/04 10:19 PM, "Tim Behrens" <[log in to unmask]> wrote: > > > Hi there - > > > > I'm guessing here, but I think it's possible you're > > thresholding the wrong image. The threshold you type in is applied to > > whichever image you have selected in the bottom left hand corner of > > fslview. My guess is that you have selected the underlying structural > > image and not the functional overlay. If you swap the selection, and then > > apply the threshold to the overlay, it should threshold ok. > > > > Tim > > > > > ------------------------------------------------------------------------------> > - > > Tim Behrens > > Centre for Functional MRI of the Brain > > The John Radcliffe Hospital > > Headley Way Oxford OX3 9DU > > Oxford University > > Work 01865 222782 > > Mobile 07980 884537 > > > ------------------------------------------------------------------------------> > - > > > > On Mon, 23 Aug 2004, Jacquie H wrote: > > > >> Hello - > >> I'm new to fsl, and need a little help with a data-viewing inconsistency. > >> I ran an analysis, and by looking at the FEAT report, could see that > >> activations in my contrasts did > >> not survive the clustered threshold analysis. When I view the data in AFNI, > >> and threshold the > >> image at 1.522, there are some nice (small) clusters of activation that I > >> would like to use for an > >> ROI analysis. > >> Here's the problem: when I load the same .hdr file (it's a tstat image) into > >> fsl view and try to > >> threshold the image, there are enormous blobs of activation everywhere - the > >> whole brain is > >> almost red (even with ridiculously high thresholds). > >> I'm wondering if anyone has encountered this problem before and could give me > >> a little advice on > >> getting the fslview images to more closely resemble the AFNI ones... > >> > >> Thanks! > >> Jacquie > >> > > > Stephen M. Smith DPhil Associate Director, FMRIB and Analysis Research Coordinator Oxford University Centre for Functional MRI of the Brain John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve