Surface-based Functional Connectivity

Hello NeuroStars Community,

I’m attempting to create a streamlined pipeline for surface-based functional connectivity from DCM to z-score, and I’m hoping for your input. Right now, I’m using fmriprep and xcp_d to go from DCM to pre-processed rs-fMRI data. This has been immensely helpful for creating correlation matrices from rs-fMRI data and standard volumetric & surface atlases.

There are many methods for seed-based functional connectivity (e.g., fsl, freesurfer, and/or xcpEngine) but I have not come across a tool for surface-based function connectivity apart from freesurfer. This is problematic because I would like to keep my files in BIDs format and/or use the CIFTI files from xcp_d that are already in fsLR_den-91k.

Is there a method for surface-based functional connectivity from xcp_d? Is it as simple as creating
a *.dlabel.nii for all the surface ROIs I plan to use in my analysis? Now that I’m thinking about it, this is probably the best method.

What surface-based functional connectivity tools do you use in your analyses and why?


Hi @bramdiamond,

Nilearn has a tutorial for seed-based surface connectivity (Seed-based connectivity on the surface - Nilearn). I have not worked through it, so I cannot say how well it would work for CIFTIs. Perhaps following this tutorial with some of the CIFTI tips mentioned in this tutorial will help. Definitely looking forward to seeing others’ thoughts on this.


1 Like