[FieldTrip] Parcellation Human connectome project-eLORETA

mehdy dousty mehdy.dousty at gmail.com
Thu Nov 3 21:14:51 CET 2016


Hello,
Thanks for the answer and sorry for my vague explanation. here is my code
to compute the inverse problem by eLORETA using the provided MRI, DATA and
 Source model in HCP.

% source localization for resting state HCP.
load('100307_MEG_3-Restin_rmegpreproc.mat') ; % loading the data;
load('100307_MEG_anatomy_headmodel.mat');% loading the headmodel
tmp = load('100307_MEG_anatomy_sourcemodel_3d6mm.mat');% sourcemodel by 6mm
individual_sourcemodel3d = tmp.sourcemodel3d;

%% MRI
individual_mri = ft_read_mri('T1w_acpc_dc_restore.nii.gz');
hcp_read_ascii('100307_MEG_anatomy_transform.txt');
individual_mri.transform = transform.vox07mm2bti;
individual_mri.coordsys = 'bti';

%% converting to the same coordination unit
individual_mri = ft_convert_units(individual_mri,'mm');
individual_sourcemodel3d = ft_convert_units(individual_sourcemodel3d,'mm');
headmodel = ft_convert_units(headmodel,'mm');
data.grad = ft_convert_units(data.grad,'mm');


%% leadfield matrix
cfg = [];
cfg.grid = individual_sourcemodel3d;
cfg.headmodel = headmodel;
cfg.grad = data.grad;
cfg.channel = ft_channelselection('MEG',data.label);
cfg.reducerank = 'no';
leadfield = ft_prepare_leadfield(cfg);


%% timelcok
cfg = [];
cfg.covariance = 'yes';
cfg.covariancewindow = [0 1000];
cfg.keeptrials = 'yes';
timelockanalaysis = ft_timelockanalysis(cfg,data);

%% Eloreta
cfg = [];
cfg.method = 'eloreta';
cfg.vol = headmodel;
cfg.grid = leadfield;
cfg.eloreta.lambda = 0.05; %Regularization parameters,cross-validation can
be
%used but as it resting state and we really dont know what the outupt looks
%like then we have to use emprical numbers which
cfg.mne.projectnoise = 'yes';
cfg.keepmom = 'yes'; %keep dipole moment
cfg.mne.keepmom = 'yes';
cfg.senstype = 'meg';
cfg.keepfilter = 'yes';
cfg.eloreta.reducerank = 'no';
source_eloreta = ft_sourceanalysis (cfg, timelockanalaysis);

so far , I compute the eLORETA, now I would like to project the data into
3d surface, like the way it was done for beamformer in
http://www.fieldtriptoolbox.org/tutorial/plotting' and then make a
parcellation based on some predefined atlases and then do the other
analysis.
I do appreciate for your helping.
Thanks



On Thu, Nov 3, 2016 at 1:53 PM Schoffelen, J.M. (Jan Mathijs) <
jan.schoffelen at donders.ru.nl> wrote:

> Hi Mehdy,
>
> It is not clear to me what you want to achieve. It’s unclear what you mean
> with ‘the results’ to be visualized on ‘the surface’, and that you have no
> ‘anatomical information’ saved in ‘the matrix’, and that there is a lack of
> ‘any mapping’. All terms between the quotation signs (for the readers among
> us who understand Dutch: my daughter aptly calls these things
> ‘bovenkomma’tjes’) are not defined precisely enough by you to give pointers
> about what or whether anything is missing, or whether something goes wrong.
>
> In general, source-level data can be parcellated with ft_sourceparcellate,
> but only if your atlas is in the same space as your functional data. That
> is, there should be a one-to-one mapping between the source locations in
> your functional data, and the source locations in your atlas. If you want
> to use the AAL atlas, which is essentially defined as a volumetric image
> (probably at a voxel resolution of 1 or 2 mm), you need to
> interpolate/downsample this atlas onto your sourcemodel at the appropriate
> resolution .This would make most sense if your sourcemodel is also defined
> as a 3D grid, but it is not absolutely necessary. In order to interpolate
> the atlas onto your sourcemodel, you could use ft_sourceinterpolate
> (provided both atlas and sourcemodel are defined in the same coordinate
> system). Note that from your messages on this forum and on the HCP
> discussion list it is not clear to the reader what source model you used
> for the eLORETA.
>
> There is some information on the fieldtrip wiki that illustrates how to
> parcellate source reconstructed data
> http://www.fieldtriptoolbox.org/tutorial/networkanalysis In the example,
> it uses a surface-based parcellation, and parcellates a connectivity
> matrix. The function can also parcellate univariate data (e.g. time courses
> or power spectra), either or not defined on a 3D grid.
>
> Also, the HCP software+documentation that the MEG team released, and which
> accompanies the released data, might give you some pointers on how to do
> it. You could try and adapt the code provided to your own needs.
>
> Good luck,
>
> Jan-Mathijs
>
>
> J.M.Schoffelen
> Senior Researcher
> Donders Centre for Cognitive Neuroimaging, Nijmegen, The Netherlands
>
>
>
> > On 03 Nov 2016, at 20:11, mehdy dousty <mehdy.dousty at gmail.com> wrote:
> >
> > Hello all,
> > By using the data of HCP, and conduct eLORETA to compute the inverse
> problem, right now Id like to visualise the results to surface but  due to
> not having any anatomical information saved in the matrix it does not show
> any mapping. Moreover, Id like to parcellate the cortex based on the AAL
> atlas, so I really appreciate if anybody can help me.
> > Thanks
> > Mehdy
> > _______________________________________________
> > fieldtrip mailing list
> > fieldtrip at donders.ru.nl
> > https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20161103/e941d860/attachment.html>


More information about the fieldtrip mailing list