yes, you are doing it wrong.
first run temporal-concat melodic. pick the IC you want and identify it in the melodic_IC. split the melodic_IC into its components using fslsplit. lets say you want to look only at the second component - use vol001 (!, not vol001 because the first will be vol000) from the melodic_IC split. feed this into the DR script as the group_IC_map. for the subjects you need to specify the preprocessed functional data of each subject, not any single-subject ICs.
does that makes sense?
hth,
andreas
________________________________________
Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag von [log in to unmask] [[log in to unmask]]
Gesendet: Freitag, 4. Februar 2011 22:10
An: [log in to unmask]
Betreff: [FSL] R: Re: [FSL] AW: [FSL] Dual regression on single component
Thanks Mikal,
I'm able to split a single subject now in all its component.
To automate the process for all my 40 subjects, i try to modify
dual_regression script in order to block it before sorting maps an running
randomise (infact i note that the scripts during the dual regression generate
the splitted maps for each component per each subject) . So I obtain all the
split component for every subject of my study.
However i can't set up propely the dual_regression for that maps. I try this
sintax:
/dual_regression 1 dr_stage2_subject00000_ic0029.nii.gz
dr_stage2_subject00001_ic0029.nii.gz dr_stage2_subject00002_ic0029.nii.gz
[............] dr_stage2_subject00039_ic0029.nii.gz dg.mat dg.con 5000 output
I wonder if the problem is the lack of <group_IC_maps> parameters...
Thanks a lot
Regards.
Nicolò
>----Messaggio originale----
>Da: [log in to unmask]
>Data: 04/02/2011 17.54
>A: <[log in to unmask]>
>Ogg: Re: [FSL] AW: [FSL] Dual regression on single component
>
>Hi Nicolò,
>
>You may use fslsplit in command line :
>
>fslsplit melodic_IC output -t
>
>it produce n files named output1, output2... correponding to your
>different components.
>
>Best regards,
>
>Mikael
>
>Le 04/02/2011 15:11, Chicco a écrit :
>> Thanks for the very fast answer.
>>
>> Can I use the fsl_regfilt command to split my data automatically?
>>
>> I already run the multi session temporal concatenation on my 40 fmri
>> session.
>>
>> Thanks a lot.
>>
>> Nicolò
>>
>>
>>
>> Il 04/02/2011 12:11, Andreas Bartsch ha scritto:
>>> Hi,
>>>
>>> sure. Just split your melodic_IC and use the IC-of-your-interest only.
>>> Cheers-
>>> Andreas
>>>
>>> ________________________________________
>>> Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag
>>> von Nicolò Cardobi [[log in to unmask]]
>>> Gesendet: Freitag, 4. Februar 2011 12:06
>>> An: [log in to unmask]
>>> Betreff: [FSL] Dual regression on single component
>>>
>>> Goodmorning,
>>>
>>> i was wondering if i can run the dual regression command on only one
>>> component (in my case the DMN one), in order to reduce the
>>> elaboration time.
>>>
>>> Cold I use the fsl_regfilt command at once on my all input?
>>>
>>> Thanks a lot and best regards.
>>>
>>> -----
>>> Nessun virus nel messaggio.
>>> Controllato da AVG - www.avg.com
>>> Versione: 10.0.1204 / Database dei virus: 1435/3421 - Data di
>>> rilascio: 03/02/2011
>>>
>>>
>>>
>>
>>
>
|