Hi Freesurfer Mailing list,

I'm examining striatal-to-cortical connections using resting state MRI and diffusion weighted imaging with tractography.

I figured the Choi and Yeo atlases in Freesurfer would be the best choice for this but I'm having some difficulties implementing them.

Since I'm curious about the anatomical specificity of FC and tracts grown, I took the conservative versions of the Choi and Yeo atlases and split them into left-and right hemispheres, and also separated those with multiple regions in a network (i.e. head of caudate and putamen tail in Choi or MFG, IPL, and SMA in one network in Yeo) with some clustering software.

The clusters match up the to original version of the conservative (Tight on the website) well and I intend to use that for FC analyses, but I would prefer to use the liberal mask for tractography.

Problematically, I've noticed that trying to separate the regions with the same clustering software on the liberal mask does not separate the regions the same way it did in the conservative mask (i.e. one cluster may contain the head of the caudate and part of the putamen).

I probably shouldn't be surprised by this, but I was wondering if you would mind explaining the way the liberal versions of these masks were created. or a reasonable alternative, so that I might be able to do the same on the parcellations I've separated out into ROIs?

I could do simple dilation I suppose, but I would like it to be as close to the original liberal mask as possible.