[FieldTrip] Looking up for right MNI coords and label

[Laura Rueda Delgado]

Dear all,

I've estimated the sources of two conditions using individual MRIs with warped grid from a template (1cm spacing). I ran a cluster-based permutation to find the significant voxels and I'm choosing the voxel where the difference is bigger. I've had some problems defining what that voxel is. I've used two options: giving the position to the atlas_lookup function, and giving the index to an interpolated tissue matrix from an atlas. Given that grid occupies the brain volume (in the BEM model with 3 layers), I also created a mask to restrict the voxels of interest to the cortex from the atlas. I seem to get different results with these and I don't understand why. All matrices are in cm. This is part of the code:

Note: The variables are
atlas = ft_read_atlas('...\fieldtrip-20141023\template\atlas\aal\ROI_MNI_V4.nii'); % then converted to 'cm'
sourcemodel = grid with 1cm spacing from template
stat = structure from ft_sourcestatistics

% Create mask from atlas
cfg = [];
cfg.interpmethod = 'nearest';
cfg.parameter = 'tissue';
sourcemodel2 = ft_sourceinterpolate(cfg,atlas,sourcemodel);
atlas_one= sourcemodel2.tissue > 0;  % Logical matrix with points with an anatomical label

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 1)Look for labels with atlas mask
[val ind_max] = max(stat.stat(:).* atlas_one(:));

% 1.a)--------------------------------------
% Atlas_lookup

pos = stat.pos(ind_max,:);
atlas_lookup(atlas, pos, 'inputcoord', 'mni')
% Results
% MNI coordinates: 4 -1 2
% Label: Rolandic_Oper_R

% 1.b)--------------------------------------
% Interpolated tissue from atlas

[xi yi zi] = ind2sub(stat.dim, ind_max);
atlas.tissuelabel{sourcemodel2.tissue(xi, yi, zi)}

% Results
% Voxel coordinates: 12 10 10
% Label: Rolandic_Oper_R

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 2)Look for labels without atlas mask
[val ind_max] = max(stat.stat(:));

% 2.a)--------------------------------------
% Atlas_lookup

% Results
% MNI coordinates: 4 -1 3
% Label: Postcentral_R

% 2.b)--------------------------------------
% Interpolated tissue from atlas

% Results
% Voxel coordinates: 12 10 11
% Label: None


What is it that I'm missing?

Thank you in advance for any help!


Best regards,



Laura Rueda Delgado
PhD student
Department of Kinesiology- Motor Control and Neural Plasticity Research Group
KU Leuven
Tervuursevest 101 bus 1501
3001 Leuven, Belgium
tel. +32 16 37 64 78


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20150310/13e254b6/attachment.html>