I would wholeheartedly second this. There is no alternative to
having a 3-D mental model of the anatomy, and the only way you can
get there is spending lots of time with atlases like the duvernoy.
what I often suggest for new students when they run their first study
is to sit down with their list of activated clusters, and go through
for each one looking at the activation on a template while cross-
referencing with the duvernoy or mai et al. atlas. I might suggest
that they have the AAL or cytoarchitectonic images in another window
as well as a cross-reference.
cheers
russ
On Jan 8, 2006, at 6:05 AM, Nicole David wrote:
> Hi Wei,
> I would strongly recommend against using the Talairach Demon to
> localize
> your functional clusters anatomically. I usually try to use several
> sources. Instead of the demon you couse use the Simon's anatomy
> toolbox
> or the aal and brodman maps as implemneted in e.g. mricro, in addition
> and more importantly you should double-check with a real atlas like
> the
> Duvernoy atlas of the human brain (on several planes, like axial and
> sagittal).
> That's how I would do it.
> Nicole
>
> Am Jan 05, 2006 10:19 PM schrieb Wei Wang <[log in to unmask]>:
>
>> Hello all,
>>
>> I have three questions related to find anatomical information.
>>
>> 1) Which function can generate mask.img and mask.hdr? I need two
>> masks
>> for both unnormalized images and normalized images.
>>
>> 2) How to obtain the anatomical information (talairach daemon)? Now I
>> am
>> doing for the following way.
>>
>> a) Normalized images. MNI coordinate -> Tal coordinate (by mni2tal)
>> then
>> search talairach daemon database.
>>
>> b) Nonnormalized iamges. I have a mask.img for nonnormalized images.
>> For
>> each voxel in the mask I set a non-zero value and set 0 for all other
>> voxels. Then do the normalization by spm_write_sn and get a
>> normalized
>> mask image. Because normalization will change the voxel size from
>> 3.125*3.125*6 to 2*2*2, there would be more than one position with
>> non-zero value. Averaging the non-zero positions let me have a MNI
>> coordinate for this voxel in nonnormalized image. So on so forth
>> to get
>> all the MNI coordinates for nonnormalized mask.
>>
>> Does this method work fine or there is another smart way to do it?
>>
>> 3) Does every coordinate in Tal space have the record in Talairach
>> Daemon database? Below are my search results, the first three numbers
>> are coordinates in Tal space.
>> -1, -71, -38, ** No data at this point. **
>> 3, -71, -38, ** No data at this point. **
>>
>> Thanks.
>>
>> Wei
>
>
> ------------------------------------
> Dipl.-Psych. Nicole David
> Klinik und Poliklinik für Psychiatrie und Psychotherapie
> Universität zu Köln
> Kerpenerstr. 62
> 50924 Köln
> Deutschland
> Tel:+49-221-478-87146
> Fax::+49-221-478-87702
> http://www.medizin.uni-koeln.de/kliniken/psychiatrie/Bildgebung/
> ------------------------------------
>
---
Russell A. Poldrack, Ph.d.
Assistant Professor
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]
web: www.poldracklab.org
|