Further, as I am really puzzeled by the change of sign (in the x voxel size)
during normalisation process, I have calculated det(Affine(1:3,1:3)) by
loading the *_sn3d.mat file.
The answer for this is negative, and going by the email sent by John
Ashburner on the 5th Dec 2001, it would indicate the images were flipped
during normalisation process. Am I right ?
Anusha
On Thu, 8 Sep 2005 09:19:44 +1000, Anusha Sritharan
<[log in to unmask]> wrote:
>Dear Volkmar,
>
>I take your point and stick to one software for the whole analysis.
>However, the FA map which has '+' x voxel size, after normalisation has
>'-' x voxel size when displayed using the same display setting.
>
>If there is no flip happening during the normalisation procedure, what
>would be the reason for the change in sign ?
>
>
>Anusha
>
>
>
>
>> Dear Anusha,
>>
>> this +/-x issue is tricky when you want to exchange data between FSL and
>> SPM. There is no explicit flip happening at some stage in SPM
>> preprocessing, instead all images are flipped for display if
>> defaults.analyze.flip is set to 1. This is of course not honoured by FSL,
>> and analyze header information is interpreted differently in FSL and SPM.
>>
>> If you don't want to leave SPM for DTI processing, you could try the
>> Diffusion II toolbox from the SPM extensions page. This way you would not
>> have to bother with image orientation issues.
>>
>> Yours
>>
>> --
>> Volkmar Glauche
>> -
>> Department of Neurology [log in to unmask]
>> Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
>> Breisacher Str. 64 Fax 49(0)761-270-5416
>> 79106 Freiburg
>>
>=========================================================================
|