If you check the initial alignment of the images using Check Reg, are
they almost aligned? If the relative displacement or rotation is too
large, then there could be instabilities.
Best regards,
-John
On 25 March 2011 11:00, Gael Varoquaux <[log in to unmask]> wrote:
> Hi list,
>
> I am getting a failure because of ill-conditionned matrices in Segment
> when I try to run it via a script, whereas through the user interface, it
> works beautifully. I am trying to figure out what the differences are,
> given that in the same Matlab window, the GUI works, and not the scripted
> job manager. As far as I can tell, I have the same options, but that's
> where the error might lie.
>
> My script is the following:
>
> ________________________________________________________________________________
> spm('Defaults','fMRI');
> spm_jobman('initcfg');
> jobs{1}.spatial{1}.preproc(1).data = {...
> '/volatile/varoquau/tmp/baronnet/spm8_epi/_subject_id_134/segment/wrsIRMAS_Patient078-0054-00001-000176-01.img,1';...
> };
> jobs{1}.spatial{1}.preproc(1).opts.regtype = 'none';
> spm_jobman('run',jobs);
> ________________________________________________________________________________
>
>
> The errors that I am getting stem from the following warning:
> ________________________________________________________________________________
> Warning: Matrix is singular, close to singular or badly scaled.
> Results may be inaccurate. RCOND = NaN.
>> In expm>PadeApproximantOfDegree at 123
> In expm at 39
> In spm_maff>P2M at 191
> In spm_maff>affreg at 152
> In spm_maff at 27
> In spm_preproc at 152
> In spm_run_preproc at 19
> In matlabbatch/private/cfg_run_cm at 29
> In cfg_util>local_runcj at 1477
> In cfg_util at 803
> In spm_jobman at 206
> ________________________________________________________________________________
>
> Any help debugging this would be greatly appreciated.
>
> Thanks,
>
> Gael
>
|