The version of FSL View I have is 1.0 beta 7 (for OS X). Is there
a newer version available? Does this version contain the bug then?
If I understand you correctly, the true values are the ones reported
by FEAT, and not the ones in fslview. I can get the values in fslview
to match FEAT's output if I change the origin in the header, so if I
know to trust
the FEAT values I can just do that.
Also, it was my mistake, the offset happens with both images.
Thanks,
Jonas
On Jul 14, 2004, at 12:16 AM, Stephen Smith wrote:
> Hi - i understand what's happening here _partly_ : there is a bug in
> the
> old version of fslview which means that Talairach coords get reported
> with
> a slight offset compared with the true values (eg what get reported in
> the
> FEAT tables). This is fixed in the new version.
>
> However, if bg_image and avg152T1 have the same header info then I
> would
> expect any repoting from fslview to be identical - is it not?
>
> Cheers.
>
>
>
> On Thu, 8 Jul 2004, Jonas Kaplan wrote:
>
>> Just a follow-up on this...
>>
>> In a group analysis, the bg_image created seems to have different
>> coordinates from the
>> avg152T1 atlas which it was registered to. That is, when functional
>> data is overlaid on the
>> bg_image the peaks come out off by 2 mm in the X direction, 4 in the y
>> direction, and 2 in the
>> Z direction relative to what is reported in the cluster list.
>> However,
>> when the functional data
>> is overlayed on the avg152 atlas, the coordinates from the list match
>> where the peaks are. Why
>> would the bg_image have different coordinates if everything is
>> registered to the standard
>> space during a group analysis?
>>
>> Thanks,
>>
>> Jonas
>>
>>
>> On Jun 25, 2004, at 7:33 AM, Stephen Smith wrote:
>>
>>> Hi - they should correspond in general - if you look at a first-level
>>> analysis and look at the non-Talairach-space version of the cluster
>>> table,
>>> the voxel positions should correspond to the voxel positions shown in
>>> FSLView. Does this work for you? The only minor issue might be that
>>> FSLView is about half a voxel out when showing standard-space images
>>> and
>>> reporting mm co-ordinates - that'll get fixed in the next version.
>>>
>>> Cheers, Steve.
>>>
>>>
>>>
>>> On Fri, 18 Jun 2004, lucina wrote:
>>>
>>>> Hi, I have a question about FSL view: In trying to locate the peak
>>>> voxels, I used the local maxima table that you can find in the
>>>> report.html for each contrast. However, it seems that the
>>>> coordinates
>>>> listed there don't match up with the actual peaks I can see in FSL
>>>> view. In other words, the coordinates listed in the tables don't
>>>> correspond to the peak activations in FSL view. Any ideas? Thanks,
>>>>
>>>> Lucina
>>>>
>>>> --
>>>> Lucina Q. Uddin
>>>> UCLA Psychology-Cognitive Neuroscience
>>>> Box 951563, B627 Franz Hall
>>>> Los Angeles, CA 90095-1563
>>>>
>>>
>>> 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
>>>
>>>
>>
>> --
>>
>> Jonas Kaplan , Ph.D.
>> Ahmanson-Lovelace Brain Mapping Center
>> University of California, Los Angeles
>> Los Angeles, CA 90095-1553
>>
>
> 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
>
|