Hi - your example image looks pretty skewed and very blurry - I would
not recommend proceeding further without resolving the registrations.
The first thing to do is to resolve whether the nonlinear
registrations are working ok - if you look in the FAi directory you
can check how well the initial registrations to the target did.
Cheers, Steve.
On 17 Jul 2007, at 23:04, Naama Barnea-Goraly wrote:
>
>
>
>> Dear Gwenaelle and FSL experts,
>>
>> Thank you for your reply. Flirt does work well on my images and
>> the output looks good with the correct dimensions. However the
>> path from there is unfortunately still bumpy.
>> The problems that followed are probably due to my lack of
>> experience so sorry about the basic question:
>>
>> After having the "flirted" FA images (using parameters from the
>> transformation of the chosen target image to the MNI template), I
>> still need to merge them, create an All_FA image, a Mean_FA image
>> and the skeleton. I edited tbss_3 with the hope of doing that (see
>> below), but my mean image looks blurry (see attached). Could you
>> please tell me what I did wrong?
>>
>> Thanks,
>>
>> Naama
>>
>>
>> #!/bin/sh
>>
>> echo [`date`] [`hostname`] [`uname -a`] [`pwd`] [$0 $@] >> .tbsslog
>>
>>
>> echo "merging all upsampled FA images into single 4D image"
>> export FSLOUTPUTTYPE=NIFTI_GZ
>> ${FSLDIR}/bin/avwmerge -t ./all_FA `$FSLDIR/bin/imglob -
>> oneperimage *flirt_*`
>> #cd ../stats
>> $FSLDIR/bin/avwcpgeom MNI152 all_FA -d
>>
>> # create mean FA and skeleton and skeleton-mask
>> echo "creating mean FA"
>> ${FSLDIR}/bin/avwmaths all_FA -Tmean mean_FA
>> echo "skeletonising mean FA"
>> ${FSLDIR}/bin/tbss_skeleton -i mean_FA -o mean_FA_skeleton
>>
>> echo "now view mean_FA_skeleton to check whether the default
>> threshold of 2000 needs changing, when running:"
>> echo "tbss_4_prestats [threshold]"
>>
>>
>>
>>
>> <mean_FA.jpg>
>>
>> On Jul 10, 2007, at 11:20 AM, Gwenaëlle DOUAUD wrote:
>>
>>> Dear Naama,
>>>
>>> sorry I have not followed all the emails about this
>>> transformation problem but it seems that everything is
>>> coming from the misregistration of your "best" target
>>> onto the MNI152. Have you identified this former and
>>> try to see if you can obtain something better just
>>> playing with Flirt and this target?
>>> Most of the time, Flirt happens to be much more robust
>>> than areg, the affine registration used to transform
>>> the best target into the MNI space in the TBSS
>>> protocol.
>>>
>>> Hope this helps,
>>> Gwenaelle
>>>
>
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|