Hi Jared
the icosahedra are only used to represent average space surface (fsaverage[4-7]), not individual subjects.
cheers Bruce On Thu, 29 Sep 2016, Jared P Zimmerman wrote:
I want to project to the native subject surfaces though, not fsaverage. If I use —trgsubject subjID would that work with the —icoorder flag? Alternatively I guess I could project to fsaverage and then use mri_surf2surf to move back to the subject surface, but that seems like a roundabout way to do it and will probably add some unnecessary interpolation. Also, I’ve used mri_vol2surf without the —trgsubject or —icoorder flags and I did successfully get output with 112745 vertices (is this FS7 space? I would have expected ~164k vertices, is this with medial wall masked?) and I could try to resample that with mri_surf2surf. I wasn’t able to do this before, but I didn’t use the flag —trgsubject or —trgicoorder flags, though I suspect I will run into the same problem trying doing this. I’m not at my computer now but can try this later this afternoon though.
Thanks, Jared
On Sep 29, 2016, at 12:00 AM, Douglas Greve <greve@nmr.mgh.harvard.edu> wrote:you can actually skip the ico thing and just use --trgsubject fsaverage6
On 9/28/16 7:06 PM, Jared P Zimmerman wrote: Thanks Doug, This does seem like it will do the right thing, but it’s failing because it’s looking for /SUBJECTS_DIR/ico/ and I do not seem to have an ico subject in my subjects_dir
Should this normally be in the subjects_dir? I don’t recall ever seeing this made when I’ve run recon-all in the past.
Thanks, Jared
Here’s the full terminal output when I run the command:
mri_vol2surf --mov /func_test/filtered_func_residuals.nii.gz --reg func_test/coreg/fs_ep2struct_fsl.dat --hemi lh --o func_test/surface/lh_surf_filtered_func_residuals.mgh --projfrac 0.5 --interp trilinear --noreshape --trgsubject ico --icoorder 6 --out_type mgh --surf-fwhm 6 IcoOrder = 6, nIcoVtxs = 40962 srcvol = /func_test/filtered_func_residuals.nii.gz srcreg = func_test/coreg/fs_ep2struct_fsl.dat srcregold = 0 srcwarp unspecified surf = white hemi = lh trgsubject = ico surfreg = sphere.reg ProjFrac = 0.5 thickness = thickness reshape = 0 interp = trilinear float2int = round GetProjMax = 0 INFO: float2int code = 0 Done loading volume Reading surface /subjdir_test//subID/surf/lh.white Done reading source surface Reading thickness /subjdir_test//subID/surf/lh.thickness Done Mapping Source Volume onto Source Subject Surface 1 0.5 0.5 0.5 using old Done mapping volume to surface Number of source voxels hit = 11345 Reading source surface registration subjdir_test//subID/surf/lh.sphere.reg Done loading source registration surface Reading icosahedron, order = 6, radius = 100 Reading icosahedron /share/apps/freesurfer/5.3.0/lib/bem/ic6.tri Done loading target registration surface Mapping Surfaces (subID -> ico) surf2surf_nnfr: building source hash (res=16). Surf2Surf: Forward Loop (40962)
surf2surf_nnfr: building target hash (res=16). Surf2Surf: Reverse Loop (112745) Reverse Loop had 72236 hits
On Sep 28, 2016, at 3:34 PM, Douglas Greve <greve@nmr.mgh.harvard.edu> wrote:You need to specify --trgsubject ico in mri_vol2surf
On 9/28/16 3:01 PM, Jared Zimmerman wrote: Hi All, forwarding this message along again as I'm still having some trouble getting this to work. Basically I would like to 1) project processed residual time-series from the subject-space volume to the subject-space surface, 2) smooth on the surface, 3) down-sample to FS4 space.
To do this is have used bbregister to calculate the registration, then mri_vol2surf to project to the surface, but I am not sure the output of mri_vol2surf is what I should be expecting. The output file has dimensions 112745 x 1 x 1 x 120 when I'm expecting ~40k vertices for FS6 space and ~3k vertices for FS4 space, but changing the --icoorder flag seems to have no effect on the number of vertices in the output. I have loaded the output surface functional image into Matlab to se that there is real time-series data in there, but I have no way of knowing if it has mapped the data appropriately. Additionally, I've tried to use mri_surf2surf to downsample this data but have not ben able to successfully get that to downsample the surface. The mri_surf2surf call I used is as follows: mri_surf2surf --srcsubject subjID --srcsurfval /mov_dir/filtered_func_residuals_fs6_lh_6mm_noreshape.mgh --trgsubject subjID --trgsurfval /mov_dir/filtered_func_residuals_fs4_lh_6mm_noreshape.mgh --hemi lh --trgicoorder 4 Any advice on how to achieve my goal of projecting to subject surfaces, smoothing and downsampling to FS4, as well as recommendations on how to visualize the surface functional data and confirm that the projections has worked properly would be much appreciated. Additionally, what should the dimensions of a surface time series look like? I would have expected for have an nVertices x nFrames x 1 array instead of the nVertices x 1 x 1 x nFrames array, is this correct? Thank you, Jared ----- Forwarded Message ----- From: "Jared Zimmerman" <jaredz@mail.med.upenn.edu> To: freesurfer@nmr.mgh.harvard.edu Sent: Monday, September 19, 2016 8:46:17 PM Subject: [Freesurfer] Projecting processed 4D functional data to individual surfaces Hi All, I'm trying to project some processed resting-state data onto the surface to run some surface based parcellations, and I'm running into a bit of difficulty. Basically I'm using bbregister to register my functional to the T1, then mri_vol2surf to project to the surface with the following pseudocode: bbregister --s subjID --mov /mov_dir/subjID_example_func_brain.nii.gz \ --reg /mov_dir/coreg/fs_ep2struct_fsl.dat \ --init-fsl --bold mri_vol2surf --src /mov_dir/filtered_func_residuals.nii.gz \ --out /mov_dir/filtered_func_residuals_fs6_lh_6mm_noreshape.mgh \ --srcreg /coreg/fs_ep2struct_fsl.dat --hemi lh \ --surf white --projfrac 0.5 --icoorder 6 --fwhm 6 --out_type mgh --noreshape The registration looks pretty good after I run bbregister, but what I get out from the mri_vol2surf seems weird. First of all, the dimensions are 112745 x 1 x 1 x 120, which seems odd. I'm forcing no reshaping, and from the help page it seems like the x-dim should be 40962 for icoorder=6. I have 120 frames in my input data, so that seems to be right, but I'm not sure where this dimension is coming from. I've also tried it with icoorder=4 and I get the same x-dim=112745 when I'm expecting 2562. I've also run these commands with various iterations e.g. reshaping, no smoothing, no projfrac, and I get the same thing every time. I'm also wondering how I should view these files to confirm that the surface projection has worked as I expect it if/when I am able to get the dimensions to be correct. I cannot open any of the projected functional files in either freeview or tksurfer, so it will be helpful to know how I might be able to view these time-series on the surface after I've succeeded in projecting them to that space. Any advice would be much appreciated Thanks, Jared _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.