Dear Bruce,
I did run it again with option (-dist 0.25); however the result did not change much (See the attachment). These results do not seem right to me and I think I’m running the command correctly (see the first email). However, I’m not sure I’m visualizing the curvature maps correctly. I assume the lh.sphere.reg has the same vertices and facets as lh.sphere but slightly displaced in space to match the curvature and sulci map of the source surface to target surface. Therefore, I visualize the lh.sphere.reg by pulling the same lh.curv file from the source image. If this is wrong, please let me know how can I correctly visualize the lh.sphere.reg, otherwise I have no idea why this surface based registration produce such inaccurate results. Any comments or suggestion is greatly appreciated.
Best [cid:E8245B5C-DB64-4625-922A-B86311C7805B@cpmc.columbia.edu] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University
Alt: razlighi@gmail.commailto:razlighi@gmail.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/
On Nov 3, 2015, at 3:47 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
the -1 means that the target is a single surface and not an atlas, but the registration is still nonlinear. The variances will all be 1 so you may have to play with the weights in the energy functional. We don't do this very much and it probably defaults to quite rigid. Try reducing the weight on the metric preservation term (e.g. -dist .25) if you want it to be more nonlinear
cheers Bruce
On Tue, 3 Nov 2015, Razlighi, Qolamreza R. wrote:
Hi Guys, I read in the sidenote here (https://surfer.nmr.mgh.harvard.edu/fswiki/SurfaceRegAndTemplates) that inter-subject surface base registration using mris_register and -1 flag performs a sort of rigid registration. So I tried it between two of my subjects with the command below mris_register -1 -curv P00001639/FreeSurferClean/surf/lh.sphere P00001639/FreeSurferClean/surf/lh.sphere lh.sphere3.reg and got the results (see the attachment). It is clear that the registration output is just a shifted version of the source. Having this confirmed I want to know if there is any way to force the mris_register to perform a complete non-linear surface based registration for inter-subjects registration the same way it does for template. I have to mention that I visualize the lh.sphere3.reg using freeview and loaded the sane lh.curv on that surface. I hope I’m not doing anything stupid.
[IMAGE] Best -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: razlighi@gmail.commailto:razlighi@gmail.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto: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.
