Re: [HCP-Users] vertex-wise correspondences across subjects

[Harms, Michael]

Hi John, Oh, this is MGH (not WU-UMN) HCP data.  Yes, I would just use ‘fsaverage’ as your registration target. If you are having problems getting ‘mri_surf2surf’ to work, I’d suggest you post the simplified version of what you are trying to do to the FS list. cheers, -MH --  Michael Harms, Ph.D. ----------------------------------------------------------- Conte Center for the Neuroscience of Mental Disorders Washington University School of Medicine Department of Psychiatry, Box 8134 660 South Euclid Ave. Tel: 314-747-6173 St. Louis, MO  63110 Email: mha...@wustl.edu From: John Griffiths <j.davidgriffi...@gmail.com> Date: Saturday, February 6, 2016 at 4:19 PM To: "Glasser, Matthew" <glass...@wustl.edu> Cc: "Harms, Michael" <mha...@wustl.edu>, "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] vertex-wise correspondences across subjects I'm doing this on the mgh connectome scanner data. There didn't appear to be any freesurfer data in that release.  Ok so sounds like you think making a custom subject average is overkill, and I should just use fsaverage? If so then would you mind just checking that I've got steps 2 and 5 right; saving to .mgh from matlab/python and projecting to the common space? At present the resulting files make freeview crash, so something's not right.  2. Write patterns out to .mgh file matlab: save_mgh(pattern,outfile,eye(4)); ...and/or python: nibabel.MGHImage(pattern[:,numpy.newaxis,numpy.newaxis].astype(np.float32),numpy.eye(4)).to_filename(outfile) 5. Project the resultant files on to the group template mri_surf2surf --srcsubject subj1 --srcsurfreg sphere.reg.avg_sub --trgsubject avg_sub --trgsurfreg sphere.reg --hemi lh --sval <pattern mgh file> --tval /tmp/test_avgsub.mgh Much appreciated!  John On 6 February 2016 at 15:58, Glasser, Matthew <glass...@wustl.edu> wrote: Indeed why use the orig surfaces?  They are missing a lot of corrections and improvements. Peace, Matt. From: <hcp-users-boun...@humanconnectome.org> on behalf of "Harms, Michael" <mha...@wustl.edu> Date: Saturday, February 6, 2016 at 2:56 PM To: John Griffiths <j.davidgriffi...@gmail.com>, "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: Re: [HCP-Users] vertex-wise correspondences across subjects Hi John, Why not just use the surfaces from HCP that already have vertex-wise correspondence?  Or, why can’t you use the usual FS machinery (i.e., through ‘fsaverage’)? If you indeed need to start from the FS ?h.orig surfaces, then this is indeed more appropriate for the FS list.  Note that I think that FS’s ‘make_average_subject’ just creates average surfaces for your particular group of subjects.  I don’t think what you are attempting actually changes the registration, since you are not changing the sphere.reg files.  (To do that is much more complicated). cheers, -MH --  Michael Harms, Ph.D. ----------------------------------------------------------- Conte Center for the Neuroscience of Mental Disorders Washington University School of Medicine Department of Psychiatry, Box 8134 660 South Euclid Ave. Tel: 314-747-6173 St. Louis, MO  63110 Email: mha...@wustl.edu From: <hcp-users-boun...@humanconnectome.org> on behalf of John Griffiths <j.davidgriffi...@gmail.com> Date: Saturday, February 6, 2016 at 11:33 AM To: "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> Subject: [HCP-Users] vertex-wise correspondences across subjects Dear hcp peeps, Could you please help me with the following analysis workflow: I am doing an analysis with HCP data where I identify patterns on the ?h.orig surfaces. What I want to do now is to resample these patterns over vertices, such that I have vertex-wise correspondence across subjects. I believe this essentially requires the construction of and projection to a group template (although I'm aware this could be done with e.g. fsaverage). Here is where I'm at: cd(freesurfer dir) 1. Read ?h.orig surfaces; compute patterns (...some in-house stuff...) 2. Write patterns out to .mgh file matlab: save_mgh(pattern,outfile,eye(4)); ...and/or python: nibabel.MGHImage(pattern[:,numpy.newaxis,numpy.newaxis].astype(np.float32),numpy.eye(4)).to_filename(outfile) 3. Generate group template make_average_subject --out avgsub --sd <fs dir> -subjects subj1 subj2 subj3... Then for each subject: 4. Register to the new group template mris_register -curv subj1/surf/lh.sphere avgsub/lh.reg.template.tif subj1/surf/lh.sphere.reg.avgsub 5. Project the surface patterns identified in 1. and 2. above on to the new group template, so that the vertex values correspond across subjects, and can be averaged, statistically compared. etc. mri_surf2surf --srcsubject subj1 --srcsurfreg sphere.reg.avg_sub --trgsubject avg_sub --trgsurfreg sphere.reg --hemi lh --sval <pattern mgh file> --tval /tmp/test_avgsub.mgh However this breaks down at mri_surf2surf, and I'm not particularly confident that the earlier parts are exactly right either. Could you please show me what commands I should be using for the aims as described above? Many thanks, John (p.s. yes this is a cross-post from freesurfer list; but got no response from there, and it is regarding analysis of HCP data, so didn't seem to heinous) -- Dr. John Griffiths Post-Doctoral Research Fellow Rotman Research Institute, Baycrest Toronto, Canada and Honorary Associate School of Physics University of Sydney _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users   The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users   The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. -- Dr. John Griffiths Post-Doctoral Research Fellow Rotman Research Institute, Baycrest Toronto, Canada and Honorary Associate School of Physics University of Sydney   The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users