I'm not sure what has gone wrong. The offending line (213) does
checking of dimensions, matrices etc:
if ~all(dm==odm) || ~all(M(:)==oM(:)),
Maybe the number of dimensions in your images is a bit different from
what would be expected (eg if it is a 4D file). If this is the case,
then you could try:
if ~all(dm(1:3)==odm) || ~all(M(:)==oM(:)),
Best regards,
-John
On Mon, 2009-08-10 at 11:39 +0100, Richard Binney wrote:
> Dearest John et al.
>
> (sorry for reposting, but i think this message got lost in the depths
> of another thread)
>
> I've had a look at the list and can't find a reference to what is
> happening for me.
>
> PLease can you tell me what I have done wrong in trying to normalise
> fMRI data to MNI using DARTEL
>
> steps:
> 1) realigned and resliced EPI images within each subject
> 2)slice timing correction (i know this is not reccomended anymmore but
> i dont quite understand why and have quite a long TR so kept that step
> in)
> 3)coregistered (estimate only as per DARTEL guide) the T1 to the mean
> EPI image (genreated in realign)
> 4)used New Segment on each T1 image (done in one go with each subject
> as separate volume in single segment job - think this is OK to do,
> correct?)
> 5)generated group template using rc1 and rc2 images of each subject
> and default settings (didn't use the other rc3, 4, etc...would this
> help?)
> 6)tried to use 'normalise to MNI space' with template 6 and flow field
> and then selecting my EPI data as images to normalise
>
> 2 things:
>
> 1)the file slector won't let me select all 465 of my dynamics as you
> would for other preprocessing steps so only inputting the single image
> hoping that it would recognise the multiple volumes..will it or is
> this a bug in the file selector (I saw a reply regarding a selector
> bug but could find previous message to see what bug was)??
>
> 2)normalise to MNI failed with the follwoing message:
>
> Running 'Normalise to MNI Space'
> ** Affine registering "Template31_6" with MNI space **
> 1 7.39744 0.127795
> 2 3.12956 0.0276395
> 3 0.177594 0.0158539
> 4 0.0747908 0.0156323
> 5 0.0528847 0.0156371
> 6 0.0290208 0.0156315
> 7 0.0145358 0.0156317
> 8 0.00701687 0.0156314
> 9 0.00333472 0.0156315
> 10 0.00157345 0.0156315
> 11 0.00073873 0.0156315
> 12 0.000345959 0.0156315
> 13 0.00016349 0.0156315
> ** "u_rc1base_p3_T1_Template31" **
> ardMRH26_3_1Failed 'Normalise to MNI Space'
> Error using ==> eq
> Matrix dimensions must agree.
> In file "C:\Program Files\MATLAB\R2007b\toolbox\spm8\toolbox\DARTEL
> \spm_dartel_norm_fun.m" (v3153), function "deal_with_subject" at line
> 213.
> In file "C:\Program Files\MATLAB\R2007b\toolbox\spm8\toolbox\DARTEL
> \spm_dartel_norm_fun.m" (v3153), function "spm_dartel_norm_fun" at
> line 123.
> The following modules did not run:
> Failed: Normalise to MNI Space
>
> I'm guessing this is that my structural and functional aren't in the
> same space but i did coregister them...so what have i done wrong.
>
> Please help!!
>
> Many thanks in advance
>
> Richard
--
John Ashburner <[log in to unmask]>
