HI - thanks very much for your input on this - that all makes sense.
Indeed, if you have non-axial images it is wise to either re-orient using
avwswapdim (the easier and better approach) or change all the flirt calls
to have the wider angular search space in the scripts.
Thanks, Steve.
On Fri, 29 Nov 2002, carles wrote:
> Hi,
> the object of this message is to report a possible SIENA misfunction when
> applied on coronal brains to warn possible users.
>
> In order to test SIENA we have adquiered three 3d's of 5 healthy volunteers,
> two in the same session another one, one or two weeks later.
> The results of SIENA on some of those studies was quite disappointing since
> variations over 2% were reported.
>
> After rejecting some hypotesis related to image format, we have checked
> SIENA carefully step by sted and we have noticed that, in some cases,
> talairach masking fails because registration with talairach template is
> wrong (the brain is in 180º respect to tempalte). That could happen when
> orbits anb soft tissue under skull basis is not removed enterely by BET so
> the brain to register is practically symmetric, in terms of volume, respect
> X-Z plain.
>
> The problem is easely removed by orienting coronal images in axial order
> (with avwswapdim, for instance) before running SIENA and keeping search
> angles below 90 degrees in flirt.
>
> Yours,
>
> Carles Falcon
> University of Barcelona
>
Stephen M. Smith
Head of Image Analysis, FMRIB
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|