Hi Ted, can you use epidewarp.fsl to create the warp field (voxel shift map (VSM))? If so, you can then use bbregister and mri_vol2surf with the --vsm option to apply the warp simultaneously with determining or applying the registration. It might work with the warp field that you have (since epidewarp.fsl uses fsl to compute the VSM).
doug
On 11/15/2012 02:35 PM, Ted Satterthwaite wrote:
Hi Bruce, Doug, and FreeSurfer Team,
We have calculated coregistrations between bold timeseries and T1 images using BBR & distortion correction as implemented in the epi_reg utility in FSL5. We would like to map our functional timeseries to surface space for connectivity analyses (analagous to papers by Yeo & Buckner). Normally, without distortion correction, we would use tkregister tp convert the rigid body coregistration calculated in fsl by flirt(.mat) to a register.dat file, and then use mri_vol2surf to birng the data to surface space. However, with distortion correction, the coregistration between the epi & T1 is represented as a deformation field (_warp.nii.gz).
Is there a way for FreeSurfer to read this warp and use the information? We could of course simply do a two-step registration by moving our epi data into T1 space first using FSL and then use FreeSurfer to get it to the surface, but that would involve an extra interpolation. Are the ways to do this in one step using our already-calculated & QA'd deformation fields from FSL?
Thanks so much for your time & advice.
Best,
Ted Satterthwaite
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer