Print

Print
Hey Donald & Evgeny,

thanks for the answers. Sorry, didn't make it clear enough: I have preprocessed resting state data from one subject and I want to extract all the voxel time courses from the grey matter. After calculation of some value for each voxel, I want to create one map (nifti file?) which shows this value at the voxel coordinates.

So far I extracted the voxel data from the normalised functional images and use the segmented grey matter as a mask in order to find the coordinates of the grey matter voxels. 
I used a threshold of 0.5 to extract the grey matter voxels from the c1-file, though I'm not sure how these files are coded.
Then I extracted the voxel time courses and calculated the value. My idea was then to create a nifti-file.

I'm not sure if this is correct or if there is an easier way (maybe via GUI)?

greetings

David

2014-08-08 14:37 GMT+02:00 Evgeny Petrovskiy <[log in to unmask]>:
If these are for visualization purposes only then it would probably be easier just to draw the areas of interest on anatomical images since it does not look like there is a specialized tool for that.

But on the lowest level possible you could take a template map, load it into Matlab via low-level functions found at Mathworks (http://www.mathworks.com/matlabcentral/fileexchange/8797-tools-for-nifti-and-analyze-image  load_nii / load_nii_untouched functions ) and change it's img property. It is a simple 3d array of numbers corresponding to the type of image you used as a template (so spm_T* files contain actual T values if I'm not mistaken).

For easier 'aiming' process you could reslice some anatomical reference into the space of your map (via this function, for example http://spect.yale.edu/downloads/chg_vx.m ) and then using view_nii from previous package (or display function of SPM -> vx box of Crosshair position) you could look up array indices corresponding to particular anatomical regions.


Also you could use AAL and talairach images to 'color' whole areas of these atlases. (you could find some in spm/canonical folder - these images contain numbers of areas as listed in atlases as values of the image => you can easily find all voxels with value of you region of interest)


Best regards,
EP