Dear Shary,
a starting point might be the following review by Brett et al., "The problem of functional localization in the human brain", which appeared in Nat Rev Neurosci. in 2002 [Nat Rev Neurosci 3(3):243-9].
In addition, you might also want to have a look at our paper describing the SPM Anatomy toolbox, where some of those issues are discussed as well [Eickhoff et al., "A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data" NeuroImage (2005) 25(4):1325-35.]
I hope this helps
Simon
-----------------------------------------
Simon Eickhoff
Institut for Medicine (IME); Research Center Jülich
Leo-Brandt-Str. 5; 52425 Juelich, Germany
Phone + 49 2461-61-5219 / Fax + 49 2461-61-2820
e-mail: [log in to unmask] / [log in to unmask]
----- Original Message -----
From: "[log in to unmask]" <[log in to unmask]>
Date: Thursday, October 13, 2005 5:52 pm
Subject: Re: Proper description of anatomical space
> Dear Simon,
>
> I definitely agree with you about the conversion of MNI
> coordinates into Talairach coordinates, because I have found some
> errors throughout my many epxeriences. I simply superimposed the
> significant blobs (PET/SPECT/VBM) onto the averaged MRI/T1 and/or
> the Clin's MRI (display in SPM), which are within MNI space, and
> compared the coordinates of the blobs with the converted
> coordinates. It was clear to me that some of converted coordinates
> were wrong.
>
> My question for you: Is there any publications or reference you
> know of that indicates such problems? If you know of such
> publications, could you please let me know.
>
> Regards,
> Shary
>
> Shahryar Rafi-Tari, M.Sc.
>
> Database Consultant / Imaging Analyst
> ADNI Imaging Project Coordinator at Sunnybrook
>
> 2075 Bayview Avenue
> Sunnybrook & Women's College Health Sciences Centre
> Linda C Campbell Cognitive Neurology Research Unit
> Room B630
> Toronto, Ontario, Canada M4N 3M5
>
> E-mails: [log in to unmask]
> Tel: 416-480-6100 ext. 3281
>
>
>
> "Simon B. Eickhoff" <[log in to unmask]> wrote:
>
> Just a short addition to this comment:
>
> I would definitively also favour the option of naming the actual
> template you have used (e.g. EPI template provided by SPM2, which
> is approximately in line with the MNI space).
>
> However, I would rather disadvice you to convert your coordinates
> into "Talairach space" and report those, since obviously this
> conversion introduces another possible source of error (in
> particular since the difference between MNI space / SPM EPI space
> and Talairach space cannot correctly be accounted for by linear
> transformations but are in some respects highly non-linear.
> Consider e.g. the location of the major sulcal landmarks.).
>
> Moreover, in my opinion, the use of "Talairach", "MNI", "SPM" and
> "SPM converted to Talairach" coordinates in one paper does not
> help clarity. Thus it might be most abvisable to stick to one
> reference space (that one, which you actually used to normalise
> your data to) and tell people the correct name of that space.
>
>
>
> Best wishes
>
> Simon
>
>
>
>
> ----- Original Message -----
>
> From: Cyril Pernet <[log in to unmask]>
>
> Date: Thursday, October 13, 2005 12:22 pm
>
> Subject: Re: [SPM] Proper description of anatomical space
>
>
>
>
> > Hi Johan
> >
> > > Dear SPM mailbase,
> > > We would like help in finding the proper description of the
> > space in
> > > which coordinates are
> > > reported when the normalization procedure is based on the EPI
> > template in
> > > spm2.
> > > > Suggestions:
> > > "The x, y, z coordinates refer to the Talairach space
> > > as implemented by the MNI brain template"?
> > > (However, as the EPI template does not exactly match
> > > the MNI brain template this description in not entirely
> > > correct)
> > > > "The x, y, z coordinates refer to an approximate
> > > Talairach space"?
> > > (This lacks precision, at least in the mind of one reviewer)
> >
> > If I remember well Jesper Andersson posted some stuff about that
> > few weeks
> > ago. My humble opinion is that normalization is anyway not
> perfect
> > so the
> > term "approximate" is always true.. anyway I now personnaly
> refer
> > to the
> > template used only: e.g. data were normalized onto the EPI
> > template as
> > implemented in SPM2 (or SPM5). Then if I report coordinates in
> > Talairach
> > space, I specify how I got these coordinates (e.g. MNI2TAL M.
> > Brett's algo).
> > This way if the reader wants more info about the template he/she
> > can have a
> > look at the template in SPM (everything is commented) and I
> don't
> > need to
> > explain the approximations from MNI to Talairach and so one.
> >
> > Not sure it really help.. :-))
> > Best
> > Cyril
> >
> >
> >
> > -------------------------------------------------
> > This mail sent through IMP: http://horde.org/imp/
> >
>
>
>
> -----------------------------------------
>
> Simon Eickhoff
> Institut for Medicine (IME); Research Center Jülich
> Leo-Brandt-Str. 5; 52425 Juelich, Germany
> Phone + 49 2461-61-5219 / Fax + 49 2461-61-2820
> e-mail: [log in to unmask] / [log in to unmask]
>
>
>
>
> ---------------------------------
> Find your next car at Yahoo! Canada Autos
>