Hi Xiaochu,
You want to choose between options 1) and 2) below, but I don't know
enough about the VBM processing stream to have a useful opinion.
Choosing either would worry me, because I'm sure the atlas is used in
many steps, and I'd be concerned that there was some undone change
lurking there. Maybe not; maybe there's an environment variable one can
set, or you can simply globally replace your atlas path in your scripts.
I don't know.
But it does seem like you're not considering the option Devlin mentioned:
> it would make more sense to transform the functional results from T&T
> space to MNI, rather than vice versa
There are many ways to compare your MNI VBM results with your other T88
results. You could do as Devlin recommended. You could also register all
your subjects' MPRAGES two ways: Once to MNI, once to +tlrc. Then you
could apply the inverse of your toMNI mat, followed by adwarp to +tlrc
to get the VBM results from MNI to +tlrc. Or you could do that just to
the final group results. Lots of options.
Karen Berman's group at NIH must deal with the MNI<->T88 problem often.
Another excellent resource is Ziad Saad.
Before jumping in, this is worth some real thought.
Donna
On 10/04/2007 01:11 PM, Zhang, Xiaochu (NIH/NIDA) [F] wrote:
> Thank you very much, Donna!
> If I don't misunderstand, you think I should select the first way, i.e., after I get "GM_mod_merg" file. Is it correct?
>
> I used AFNI more than 8 years. You know, in the early version, AFNI only support Talariach space. In NIH, much more persons are using AFNI than other software. Additionally, "Where am I" is a really good feature for me. I really like it and use it usually.
>
> Thank you, again!
>
>
> -----Original Message-----
> From: Donna Dierker [mailto:[log in to unmask]]
> Sent: 2007年10月4日 8:54
> To: [log in to unmask]
> Subject: Re: [FSL] VBM and Talairach
>
> Hi Xiaochu,
>
> Poldrack and Devlin's recommendations are worth heeding, but I should
> point out another option available to you: AFNI provides the popular
> templates (e.g., ICBM152, ICBM452) shrunken down to fit the Talairach
> bounding box, and you can now use AFNI's @auto_tlrc to affine register
> your volumes to those +tlrc versions of the templates. This option at
> least gets you away from the frankenstein approach to stereotaxic
> registration, and lets you enjoy the benefits of having your data in T88
> space.
>
> To be honest, my life would be easier if everyone adopted a common
> space, as Poldrack and Devlin recommend in the first link below. But
> the AFNI folks, for example, have a great deal of legacy data in T88,
> and enjoy their "Talairach to" and "Where am I" features (not that they
> couldn't be made to work in MNI space).
>
> Like Russ, I'd like to know what is driving your T88 preference.
>
> Donna
>
> On 10/03/2007 05:55 PM, Russell Poldrack wrote:
>
>> Why do you need to transfer to Talairach space? See these papers for
>> reasons why you might not want to do that:
>>
>> http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&Cmd=ShowDetailView&TermToSearch=17870621
>> <http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&Cmd=ShowDetailView&TermToSearch=17870621>
>>
>> http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&Cmd=ShowDetailView&TermToSearch=11994756
>> <http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&Cmd=ShowDetailView&TermToSearch=11994756>
>>
>> cheers
>> russ
>>
>>
>> On Oct 3, 2007, at 3:49 PM, Zhang, Xiaochu (NIH/NIDA) [F] wrote:
>>
>>
>>> Hi FSL experts,
>>>
>>> I’m working on the VBM data with FSL-VBM. FSL is really good at the
>>> VBM analysis. Thank you very much for all FSL guys!
>>>
>>>
>>>
>>> However, I need combine the VBM data to some other kinds of data
>>> (e.g., the BOLD data) and these data had been analyzed by AFNI.
>>>
>>> The VBM is in MNI space, now. Therefore, I need transfer it to
>>> Talairach space.
>>>
>>> Now, there are two ways:
>>>
>>> 1. Ttransfer to Talairach from MNI after fslvbm_3_proc.
>>>
>>> 2. Let the grey-matter partial volume aligned to Talairach standard
>>> space instead of MNI152 space.
>>>
>>>
>>>
>>> The first question is which way is better.
>>>
>>>
>>>
>>> I prefer the second way.
>>>
>>> However, I don’t know how to modify the script (fslvbm_1_bet?
>>> Fslvbm_2_template? Or fslvbm_3_proc?) and where I can find the
>>> Talairach space template file (for replace the MNI152 file).
>>>
>>>
>>>
>>> Thank you very much!
>>>
>>> Xiaochu Zhang Ph.D
>>>
>>> Visiting Research Fellow
>>>
>>> NIH/NIDA-IRP
>>>
>>> 5500 Nathan Shock Drive
>>>
>>> Baltimore MD 21224
>>>
>>>
>>>
>>> Tel: (410) - 550 - 1440 ext. 434
>>>
>>>
>>>
>>>
>>>
>> ---
>> Russell A. Poldrack, Ph.d.
>> Associate Professor
>> Wendell Jeffrey and Bernice Wenzel Term Chair in Behavioral Neuroscience
>> UCLA Department of Psychology
>> Franz Hall, Box 951563
>> Los Angeles, CA 90095-1563
>>
>> phone: 310-794-1224
>> fax: 310-206-5895
>> email: [log in to unmask] <mailto:[log in to unmask]>
>> web: www.poldracklab.org <http://www.poldracklab.org>
>>
>>
>>
>>
>
>
|