Dear Volkmar,
right now I am implementing it exactly like that; user and study
specific spm_defaults.m files that are in a folder set to be in the
matlab path above the spm5 folder. They are copied to the working
directory for future reference.
Indeed, I would recommend to put some of the most important (if not all)
default options to be configurable from the job manager.
Thanks for sharing your ideas,
Bas
Volkmar Glauche schreef:
> Dear Bas,
>
> you are probably right, this should go into a configuration option for
> the job manager. The defaults.mask.thresh and some other options are
> currently bypassing SPM5 job management completely. A workaround would
> be to implement study specific spm_defaults.m files.
>
> Volkmar
>
> Am Dienstag, den 08.04.2008, 16:56 +0200 schrieb Neggers, S.F.W.:
>
>> Dear List,
>>
>> in spm_defaults.m, one can change settings for creating a mask of the
>> brain (to exclude analysis of out of brain voxels) using
>> defaults.mask.thresh, which defaults to 0.8. For quite a few analyses,
>> I need lower values. This is especially when using parallel imaging
>> (SENSE), that leads to pretty strong intensity gradients in an image,
>> leading to masking out of some in-brain voxels when using 0.8.
>>
>> However, what I'm looking for is an option to set this masking default
>> (and, for that matter, any other default setting from spm_defaults) in
>> the job-manager menu, for use with spm_jobman('run',myjobs). Any other
>> method not needing permanent alterations of the spm5 code would also
>> be worthwile. I looked for it but couldn't locate it in the
>> jobmanager.
>>
>> That would be much more practical than changing spm_defaults.m every
>> time, and would allow logging of what was actually done. Besides, we
>> use one central SPM5 version for the whole group, and changing
>> spm_defaults.m for one study would change it for all users, where I
>> would like to change it per study depending on the acquisition
>> technique used.
>>
>> In spm2 one could symply use
>>
>> global defaults
>> defaults.mask.thresh=<something else>;
>>
>> after launching spm2, and the altered threshold would be used.
>> In spm5, I found that spm_defaults is called every time a model is
>> specified or estimated, so the above doesn't work anymore.
>>
>> Any suggestions?
>>
>> Kind regards,
>>
>> Bas
>>
>>
>> --------------------------------------------------
>> Dr. S.F.W. Neggers
>> Division of Brain Research
>> Rudolf Magnus Institute for Neuroscience
>> Utrecht University Medical Center
>>
>> Visiting : Heidelberglaan 100, 3584 CX Utrecht
>> Room B.01.1.03
>> Mail : Huispost B.01.206, P.O. Box 85500
>> 3508 GA Utrecht, the Netherlands
>> Tel : +31 (0)88 7559609
>> Fax : +31 (0)88 7555443
>> E-mail : [log in to unmask]
>> Web : http://www.fmri.nl/people/bas.html
>> --------------------------------------------------
>>
>>
>>
>>
--
--------------------------------------------------
Dr. S.F.W. Neggers
Division of Brain Research
Rudolf Magnus Institute for Neuroscience
Utrecht University Medical Center
Visiting : Heidelberglaan 100, 3584 CX Utrecht
Room B.01.1.03
Mail : Huispost B.01.206, P.O. Box 85500
3508 GA Utrecht, the Netherlands
Tel : +31 (0)30 2509609
Fax : +31 (0)30 2505443
E-mail : [log in to unmask]
Web : http://www.fmri.nl/people/bas.html
--------------------------------------------------
|