I've got a nifty file of someone's T1 as well as an mgrid file of electrodes that have been localized to this space (from CT scans). What I'd like to do is load this nifty file into a program like freesurfer and plot the electrodes on top of it there. (ideally after doing things like surface estimation etc).

What I've done is run the freesurfer surface reconstructions on this data (which outputs its own T1), and wrote a script to go through the .mgrid file and pull the x/y/z coordinates for each electrode.

I've tried two things, neither of which work the way I want them to:

1. Use the x/y/z locations from the mgrid file to plot a 3D brain w/ electrodes on the Freesurfer T1
2. Use the x/y/z locations from the mgrid file to plot on a 3D rendering of the nifti file I used for coregistration in bioimagesuite.

In both cases, the electrode locations are skewed and translated away from the brain, so clearly they're in the wrong coordinate space, but I'm not sure what transformations to apply in order to make this work. I assume that the mgrid files store things in x/y/z space, not i/j/k/ space, but not sure how that maps on to what Freesurfer uses.

Any ideas on how to accomplish this?
Thanks

You may want to export the electrodes in .vtk format or something like that to test. The mgrid files store data in mm with (0,0,0) being the center of the first voxel in the image.