[FieldTrip] Apply atlas to MNE grid
Schoffelen, J.M. (Jan Mathijs)
jan.schoffelen at donders.ru.nl
Sun Jun 14 19:49:39 CEST 2015
a) The reason I doubt the validity of this method is that when I find high activity on the MNE grid in seemingly temporal or occipital sources, the blob on the MRI image tends to end up on the brainstem or cerebellum (an example can be seen in the attached image). Any thoughts on what could cause this? Could there for example be an issue with cerebellar grid points not being included in the MNE grid, but which are present in the beamformer/cubic grid? Do you have any thoughts on the validity of this approach and the resulting effect?
Essentially, when you are reconstructing the source activity on the cotrical sheet, but interpolate onto a 3D grid, you are basically extrapolating data onto voxels that fall outside the cortical sheet. Thus, indeed, cerebellum and brainstrem values need to be taken with a very large pinch of salt. A suggestion here would be to create a voxel-based mask, that masks out all voxels that are too far away from the cortical sheet.
b) This pipeline feels rather contrived, with a bit too many interpolations to my liking. Is it somehow possible to shorten this pipeline? For example, is it possible to directly link the vertices of the MNE surface to an atlas?
You could consider another route, where you use a surface based atlas (e.g. the Conte 69 atlas, based on Brodmann area labels). This would however require your cortical sheets to be registered in surface space (to a template). This can be done with the Caret package, or better still with some dedicated HCP code. You can google for ‘freesurfer fs to LR’ if you want to go this route and get started.
Any thoughts or hints on this would be greatly appreciated.
fieldtrip mailing list
fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the fieldtrip