Thanks for your help. The output for my first regression has finished now,
but I am unable to open the file. I don't think it worked properly. I have
done something similar to this before, but I used the timecourses from
melodic_mix instead of melodic_IC.nii.gz Do I need to extract single
volumes from melodic_IC.nii.gz for this to work? Thanks for your help.
Chris Bell
On Nov 14 2008, Christian F. Beckmann wrote:
>Hi
>
>You can cut down on memory use if you also specify the correct mask
>usign the -m option to fsl_glm.
>Also, the second regression can actually be performed against the
>filtered data in native space (if available) so that denoising can
>then be carried out in native space
>hth
>Christian
>
>
>On 14 Nov 2008, at 23:07, Christopher Bell wrote:
>
>> Christian,
>>
>> I tried this first regression step on a group dataset I had already
>> run through melodic, against a subject's 4D filtered_func_data in
>> global space 2x2x2 mm resolution, and was finding it quite memory
>> intensive. It was using up to 7GB of memory to run this application,
>> is this what you would expect?
>>
>> Chris
>>
>>
>>
>>
>>
>> On Nov 14 2008, Christian F. Beckmann wrote:
>>
>>> Hi Jim,
>>>
>>> For de-noising to work you need subject-specific spatial maps and
>>> time courses. You can get them quite easily by regressing the
>>> subject specific data sets against the melodic_IC file and then re-
>>> regressing the same data set against the output from the first
>>> regression:
>>>
>>> (i) fsl_glm -i filtered_func_X -d group_melodic_IC -o timecourses_X
>>>
>>> (ii) fsl_glm -i filtered_func_X -d timecourses_X -o maps_X
>>>
>>> After this you can simply follow the normal melodic instructions
>>> and use timecourses_X and maps_X for each data set X
>>>
>>> hth
>>> Christian
>>>
>>>
>>>
>>>
>>> On 11 Nov 2008, at 20:18, James Porter wrote:
>>>
>>>> Hello-
>>>>
>>>> On the Melodic instructions page there are clear instructions on
>>>> how to denoise a functional volume at the single-subject single-
>>>> scan level. However, if one wants to denoise a group-wise data
>>>> set after running tensor-ICA, is there an equivalent procedure?
>>>>
>>>> --
>>>> ---------
>>>> Jim Porter
>>>> Graduate Student
>>>> Clinical Science & Psychopathology Research
>>>> University of Minnesota
>>>
>>>
>>> _______________________________________________
>>> Christian F. Beckmann, DPhil
>>> Senior Lecturer, Clinical Neuroscience Department
>>> Division of Neuroscience and Mental Health
>>> Imperial College London, Hammersmith Campus
>>> Rm 419, Burlington Danes Bldg, Du Cane Road, London W12 0NN, UK
>>> Tel.: +44 (0)20 7594 6685 --- Fax: +44 (0)20 7594 6548
>>> Email: [log in to unmask]
>>> http://www.imperial.ac.uk/medicine/people/c.beckmann/
>>>
>>> Senior Research Fellow, FMRIB Centre
>>> University of Oxford
>>> JR Hospital - Oxford OX3 9DU
>>> Tel.: +44 (0)1865 222551 --- Fax: +44 (0)1865 222717
>>> Email: [log in to unmask]
>>> http://www.fmrib.ox.ac.uk/~beckmann
>>>
>
>
>_______________________________________________
>Christian F. Beckmann, DPhil
>Senior Lecturer, Clinical Neuroscience Department
>Division of Neuroscience and Mental Health
>Imperial College London, Hammersmith Campus
>Rm 419, Burlington Danes Bldg, Du Cane Road, London W12 0NN, UK
>Tel.: +44 (0)20 7594 6685 --- Fax: +44 (0)20 7594 6548
>Email: [log in to unmask]
>http://www.imperial.ac.uk/medicine/people/c.beckmann/
>
>Senior Research Fellow, FMRIB Centre
>University of Oxford
>JR Hospital - Oxford OX3 9DU
>Tel.: +44 (0)1865 222551 --- Fax: +44 (0)1865 222717
>Email: [log in to unmask]
>http://www.fmrib.ox.ac.uk/~beckmann
>
|