On Tue, Sep 2, 2014 at 5:11 AM, Stephen Smith <[email protected]> wrote:

> Hi
>
> On 2 Sep 2014, at 10:28, David Hofmann <[email protected]> wrote:
>
> > > after taking a look into the HCP related papers, I decided to download
> the FIX/pre-processed rfMRI-data for 1 subject (100307_3T_rfMRI_REST_fix).
> I want the "ready-to-analyze"-pre-processed data for the first session and
> assume the volumetric data is provided via the fix dataset and the
> following nifti-files (only first rfMRI-session):
> > >
> > > -rfMRI_REST1_LR_hp2000_clean.nii
> > > -rfMRI_REST1_RL_hp2000_clean.nii
> >
> > no - these are the "grayordinate" CIFTI files combining cortical
> vertices and noncortical voxels.
> > rfMRI_REST1_LR_hp2000_clean.nii.gz
> > etc.
> >
> > I'm sorry, I made a mistake with the names of the dataset. I used the
> fix extended dataset (100307_3T_rfMRI_REST1_fixextended). So, just to be
> sure, the nifti-files are in:
> >
> > -
> ...fixextended\MNINonLinear\Results\rfMRI_REST1_RL\rfMRI_REST1_RL_hp2000_clean.nii.gz
> > -
> ...fixextended\MNINonLinear\Results\rfMRI_REST1_LR\rfMRI_REST1_LR_hp2000_clean.nii.gz
> >
> > After unzipping the above files:
> >
> > -rfMRI_REST1_RL_hp2000_clean.nii
> > -rfMRI_REST1_LR_hp2000_clean.nii
> >
> > I assume this have to be the correct NIFTI-files for the first session.
>
> apologies for the confusion - yes, these are the volumetric files.
> The grayordinate files are differently named - e.g.
> rfMRI_REST1_LR_Atlas_hp2000_clean.dtseries.nii
>
> Btw - there's no need to unzip these volumetric NIFTI files - at least for
> FSL usage.


Workbench will also read them without manual unzipping.


>  >> > in short, I want to handle the data like a pre-processed nii-file
> ready to extract the voxel/ROI time course data.
> >> >
> >> > since I'm still a little uncertain about how to proceed and if I have
> the correct dataset the following questions arise:
> >> >
> >> > 1. Is it necessary to combine/concatenate LR and RL - data? If so, is
> there a way to combine them via the workbench?
> >>
> >> One way or another you will want to combine them, yes.  In the upcoming
> release the 4 runs for a subject are already concatenated inside the
> node-timeseries files.
> >
> > Until then, I understand I have to normalize (demean) and then combine
> the LR,RL-files and this can be done for example with Matlab by extracting
> the voxel time courses and concatenating them. Is this the correct way or
> is there an easier way by using the workbench tool?
>
> Doing that in matlab is fine, yes.   I suspect that you can do it with the
> workbench tool, but I'm not so familiar with that.  Workbench command line
> tool is more oriented towards CIFTI/grayordinate representations than
> volumetric.


Actually, most operations you can do on cifti are also available to volume
(and surface) - correlation is the main exception, as (numvoxels)^2 would
make for an absurdly large file, and a lot of unneccesary computation
(white matter, anything outside the brain).

See this thread for a previous discussion of independently normalizing each
brainordinate's timeseries using wb_command:

http://www.mail-archive.com/[email protected]/msg00444.html

-volume-reduce and -volume-math exist and behave similarly, but instead of
the "-select 1 1 -repeat" parts in the -cifti-math command, you will need
"-subvolume 1 -repeat"

 >> > 4. Is there a mask-file for the localization of voxels? I only found
> a mask-file in the 100307_3T_rfMRI_REST1_preproc dataset in which there
> seems to be no correction of motion and physiologic artefacts(?)
> >>
> >> Not sure what you mean by a mask file...
> >
> > For example like the file: brainmask_fs.2.nii.gz provided in the
> preproc-dataset in order to localize the cortical and subcortical voxel
> coordinates. I can't find such a file provied in the fix extended dataset.
>
> The volumetric datasets are already in 2mm MNI152 standard space, so
> (e.g.) atlases provided with FSL may be helpful depending on exactly what
> you're wanting to do.
> All subjects' volumetric datasets should be in the same space.
>
> Cheers.
>
> ---------------------------------------------------------------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director,  Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
> +44 (0) 1865 222726  (fax 222717)
> [email protected]    http://www.fmrib.ox.ac.uk/~steve
> ---------------------------------------------------------------------------
>
> Stop the cultural destruction of Tibet
>
>
>
>
>
> _______________________________________________
> HCP-Users mailing list
> [email protected]
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to