> From: "n.ramnani" <[log in to unmask]>
> Subject: Re: "Brain Atlas for Functional Imaging"
> Comments: To: Andreas Frank <[log in to unmask]>
>
> Im afraid this is not strictly speaking correct. There are only very
> approximate spatial relationships between the atlas of Talairach and
> Tournoux and the ICBM /MNI space to which we normalise our images. The
> atlas is now only appropriate to gain a 'very general' idea of where
> activations are. The experiences of many imagers has proved that the
> exclusive use of this atlas will often result mislocalisation. Brains in
> ICBM space that are normalised to the MNI template are basically of a
> different shape to the brain in the Talairach and Tournoux atlas. This is a
> debate that has gone on (and on and on...) for the last few years and I
> wont bore you with all the details (search for 'Talairach' in the SPM
> archives).
>
> There is no method that can localise activity with total accuracy, but one
> could be much more confident of a method that superimposes the functional
> data in ICBM space onto structural T1 data also in ICBM space, and then use
> a 'proper' anatomical atlas to localise the activations based on
> sulcal/gyral/other anatomical landmarks. A very suitable atlas is by
> Duvernoy (1991) ('The human brain : surface, three-dimensional sectional
> anatomy and MRI'; Wien : Springer-Verlag).
Just a warning to be careful when superimposing EPI based functional
data onto a T1 template. These images have inherently different
geometry, due to the way they are collected. Even after normalisation
the images will not match well in regions of high susceptibility (eg
brain stem, temporal lobes, near the fronal sinuses
...) unless large degrees of local warping are allowed (which has /
can have its own problems).
Also, just a reminder to always look at the orginal, T2* weighted
images alongside the activation maps. In regions where there are
large distortions there are also likely to be signal dropouts; since
these affect both the BOLD weighted and unweighted images, no
activation can/will be seen in these areas. (I know this is obvious,
but when looking at activations overlaid on T1 images it's easy to
forget that there are 'black holes' in which no activation is likely to
be seen, regardless of paradigm).
_________________
Gareth J. Barker, Institute of Neurology, Queen Square,
London WC1N 3BG, UK.
email : [log in to unmask]
SMS (phone/pager) : [log in to unmask] (< 100 characters, please!)
phone : +44 20 7837 3611 Xn 4269
fax : +44 20 7278 5616
>
> Best wishes,
>
> Narender Ramnani
>
> ***************************************************
> Dr Narender Ramnani
>
> Sensorimotor Control Group,
> Department of Physiology,
> University of Oxford,
> Parks Road,
> Oxford OX1 3TP,
> United Kingdom.
>
> Oxford University Centre for
> Functional Magnetic Resonance Imaging of the Brain,
> John Radcliffe Hospital,
> Headington,
> Oxford OX3 9DU,
> United Kingdom.
>
> Tel. 01865 222704 (Direct)
> 01865 222729 (Admin)
> mob. 0771 2632785
> Fax. 01865 222717
> email [log in to unmask]
> *****************************************************
>
