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.
[cid:AEB13CCB-455D-4591-B278-FEF31F6C0B4B@cpmc.columbia.edu]
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/
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.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/
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.
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University
Alt: 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.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.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.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.
Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here?
Best
On Nov 4, 2015, at 5:56 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: 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.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.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.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
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.
Hi Ray
what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small)
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here?
Best
On Nov 4, 2015, at 5:56 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: 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.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.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.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
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
Dear Bruce, Thank again for your kind help. I’m running version 5.1.0. Below is the output of the command. I also ran it with parea=0 but almost no change in the result (see the attachment-1) I also plot the sluci map for comparison (see the attachment-2) but still not satisfactory.[cid:8FA10C76-5448-4E9D-A3F4-C29354ADF5D9@cpmc.columbia.edu][cid:9DD54DBE-FC6F-433B-A2F7-54D77C2DCF0B@cpmc.columbia.edu]
Best
~/Data/P00001613/FreeSurferClean/surf> mris_register -1 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25.reg treating target as a single subject's surface... using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0
-------------------- tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 Starting MRISrigidBodyAlignGlobal() d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=0.9987 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.5086 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.0493 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.5976 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.1563 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=3.6949 MRISrigidBodyAlignGlobal() done 4.24 min
curvature mean = 0.036, std = 0.923 curvature mean = 0.010, std = 0.931 curvature mean = 0.036, std = 0.951 curvature mean = 0.004, std = 0.967 curvature mean = 0.036, std = 0.962 curvature mean = 0.001, std = 0.984 2 Reading smoothwm curvature mean = -0.027, std = 0.269 tol=1.0e+00, sigma=0.5, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization curvature mean = 0.090, std = 0.337 curvature mean = 0.059, std = 0.386 curvature mean = 0.091, std = 0.507 curvature mean = 0.020, std = 0.580 curvature mean = 0.091, std = 0.618 curvature mean = 0.011, std = 0.723 curvature mean = 0.093, std = 0.681 curvature mean = 0.004, std = 0.828 MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25.reg...
-- 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 4, 2015, at 7:30 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small)
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here?
Best
On Nov 4, 2015, at 5:56 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- 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. _______________________________________________ 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.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ 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.
Hi Ray
try doing:
setenv DIAG 0x04040
then run it again and send me the output
Bruce
On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote:
Dear Bruce, Thank again for your kind help. I’m running version 5.1.0. Below is the output of the command. I also ran it with parea=0 but almost no change in the result (see the attachment-1) I also plot the sluci map for comparison (see the attachment-2) but still not satisfactory.[IMAGE][IMAGE]
Best
~/Data/P00001613/FreeSurferClean/surf> mris_register -1 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25.reg treating target as a single subject's surface... using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0
tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 Starting MRISrigidBodyAlignGlobal() d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=0.9987 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.5086 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.0493 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.5976 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.1563 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=3.6949 MRISrigidBodyAlignGlobal() done 4.24 min
curvature mean = 0.036, std = 0.923 curvature mean = 0.010, std = 0.931 curvature mean = 0.036, std = 0.951 curvature mean = 0.004, std = 0.967 curvature mean = 0.036, std = 0.962 curvature mean = 0.001, std = 0.984 2 Reading smoothwm curvature mean = -0.027, std = 0.269 tol=1.0e+00, sigma=0.5, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization curvature mean = 0.090, std = 0.337 curvature mean = 0.059, std = 0.386 curvature mean = 0.091, std = 0.507 curvature mean = 0.020, std = 0.580 curvature mean = 0.091, std = 0.618 curvature mean = 0.011, std = 0.723 curvature mean = 0.093, std = 0.681 curvature mean = 0.004, std = 0.828 MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25.reg...
-- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University
Alt: razlighi@gmail.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/
On Nov 4, 2015, at 7:30 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Ray what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small) cheers Bruce On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote: Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here? Best On Nov 4, 2015, at 5:56 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote: Hi Ray that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps. Which inaccuracy are you referring to? cheers Bruce On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote: 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 [IMAGE] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: 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.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.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.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 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 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.
Dear Bruce, I did set the DIAG and ran it again with no luck. Below is the output and attached please find the result.
Best[cid:2C6B1DDA-86A5-49DD-BC4B-F518E975775B@cpmc.columbia.edu]
ray@athens-b8-e8-56-47-e6-54:~> tcsh [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~] ray% setenv DIAG 0x04040 [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~/Data/P00001613/FreeSurferClean] ray% cd surf/ [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% mris_register -1 -parea 0 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25_parea0.reg treating target as a single subject's surface... using l_parea = 0.000 using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0
-------------------- tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 reading precomputed curvature from lh.sulc
blurring surfaces with sigma=4.00... done. curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 finding optimal rigid alignment Starting MRISrigidBodyAlignGlobal() 000: dt: 0.000, sse: 135379.3 (0.279, 21.6, 0.412, 0.853), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 64.00 degree nbhd, min sse = 95165.92 (+64.00, +64.00, -64.00), min @ (0.00, 0.00, 0.00) = 95165.9 scanning 32.00 degree nbhd, min sse = 95165.92 (+32.00, +32.00, -32.00), min @ (8.00, 0.00, 0.00) = 90765.7 d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=1.1635 min sse = 90765.70 at (8.00, 0.00, 0.00) 001: dt: 0.000, sse: 130979.1 (0.279, 21.6, 0.412, 0.833), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 16.00 degree nbhd, min sse = 90765.70 (+16.00, +16.00, -16.00), min @ (-4.00, -4.00, 0.00) = 87371.6 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.7380 min sse = 87371.58 at (-4.00, -4.00, 0.00) 002: dt: 0.000, sse: 127584.9 (0.279, 21.6, 0.412, 0.817), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 8.00 degree nbhd, min sse = 87371.58 (+8.00, +8.00, -8.00), min @ (-2.00, 0.00, 0.00) = 86754.0 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.3264 min sse = 86754.03 at (-2.00, 0.00, 0.00) 003: dt: 0.000, sse: 126967.4 (0.279, 21.6, 0.412, 0.814), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 4.00 degree nbhd, min sse = 86754.03 (+4.00, +4.00, -4.00), min @ (1.00, 1.00, 0.00) = 86449.9 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.9298 min sse = 86449.91 at (1.00, 1.00, 0.00) 004: dt: 0.000, sse: 126663.3 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 2.00 degree nbhd, min sse = 86449.91 (+2.00, +2.00, -2.00), min @ (0.00, -0.50, 0.50) = 86436.5 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.5333 min sse = 86436.46 at (0.00, -0.50, 0.50) 005: dt: 0.000, sse: 126649.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 1.00 degree nbhd, min sse = 86436.45 (+1.00, +1.00, -1.00), min @ (0.00, 0.25, -0.25) = 86423.4 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=4.1242 min sse = 86423.41 at (0.00, 0.25, -0.25) 006: dt: 0.000, sse: 126636.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 0.50 degree nbhd, min sse = 86423.41 (+0.50, +0.50, -0.50), min @ (0.00, 0.00, 0.00) = 86423.4 MRISrigidBodyAlignGlobal() done 4.71 min tol=5.0e-01, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 007: dt: 241.975, sse: 109215.9 (0.308, 23.5, 0.429, 0.709), neg: 48 (%0.00:%0.01), avgs: 1024 008: dt: 117.533, sse: 99575.6 (0.298, 23.7, 0.432, 0.651), neg: 50 (%0.00:%0.01), avgs: 1024 009: dt: 136.508, sse: 95352.8 (0.305, 24.1, 0.436, 0.620), neg: 101 (%0.00:%0.02), avgs: 1024 010: dt: 105.985, sse: 92867.4 (0.303, 24.8, 0.442, 0.597), neg: 176 (%0.01:%0.03), avgs: 1024 011: dt: 149.343, sse: 90631.3 (0.311, 25.4, 0.448, 0.573), neg: 334 (%0.01:%0.06), avgs: 1024 012: dt: 85.796, sse: 89225.0 (0.312, 25.9, 0.453, 0.557), neg: 439 (%0.02:%0.08), avgs: 1024 013: dt: 177.745, sse: 87895.2 (0.321, 26.7, 0.460, 0.536), neg: 690 (%0.04:%0.13), avgs: 1024 014: dt: 66.370, sse: 87281.5 (0.322, 27.0, 0.463, 0.527), neg: 782 (%0.05:%0.14), avgs: 1024 015: dt: 236.440, sse: 86602.8 (0.331, 27.8, 0.472, 0.508), neg: 1155 (%0.08:%0.22), avgs: 1024 016: dt: 69.671, sse: 86419.9 (0.332, 28.0, 0.474, 0.502), neg: 1213 (%0.09:%0.23), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 017: dt: 122.246, sse: 82435.7 (0.347, 29.5, 0.492, 0.436), neg: 2026 (%0.16:%0.43), avgs: 256 018: dt: 69.015, sse: 81795.8 (0.350, 29.8, 0.499, 0.417), neg: 1971 (%0.12:%0.42), avgs: 256 019: dt: 36.896, sse: 81619.9 (0.353, 30.1, 0.502, 0.407), neg: 2059 (%0.12:%0.44), avgs: 256 integrating with navgs=64 and tol=1.260e-01 020: dt: 33.532, sse: 80387.5 (0.362, 30.7, 0.513, 0.370), neg: 2300 (%0.10:%0.49), avgs: 64 021: dt: 9.860, sse: 80325.9 (0.364, 30.9, 0.516, 0.362), neg: 2391 (%0.10:%0.51), avgs: 64 integrating with navgs=16 and tol=6.442e-02 022: dt: 6.761, sse: 80080.1 (0.368, 30.9, 0.520, 0.348), neg: 1935 (%0.07:%0.41), avgs: 16 023: dt: 0.929, sse: 80074.6 (0.368, 31.0, 0.521, 0.347), neg: 1928 (%0.06:%0.40), avgs: 16 integrating with navgs=4 and tol=3.494e-02 024: dt: 1.444, sse: 80034.9 (0.370, 31.0, 0.522, 0.343), neg: 1734 (%0.06:%0.35), avgs: 4 025: dt: 0.556, sse: 80028.6 (0.370, 31.0, 0.523, 0.341), neg: 1762 (%0.05:%0.36), avgs: 4 integrating with navgs=1 and tol=2.210e-02 026: dt: 0.066, sse: 80027.9 (0.370, 31.0, 0.523, 0.341), neg: 1753 (%0.05:%0.36), avgs: 1 integrating with navgs=0 and tol=1.562e-02 027: dt: 0.050, sse: 80016.6 (0.370, 31.0, 0.523, 0.341), neg: 1716 (%0.05:%0.33), avgs: 0
blurring surfaces with sigma=2.00... done. curvature mean = 0.035, std = 0.927 curvature mean = 0.011, std = 0.928 tol=5.0e-01, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 028: dt: 137.600, sse: 83361.7 (0.371, 31.1, 0.524, 0.373), neg: 1570 (%0.04:%0.30), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 029: dt: 16.658, sse: 83344.4 (0.372, 31.1, 0.526, 0.370), neg: 1484 (%0.03:%0.28), avgs: 256 integrating with navgs=64 and tol=1.260e-01 030: dt: 10.485, sse: 83245.6 (0.375, 31.4, 0.530, 0.359), neg: 1316 (%0.03:%0.23), avgs: 64 integrating with navgs=16 and tol=6.442e-02 031: dt: 4.712, sse: 83112.1 (0.379, 31.6, 0.534, 0.346), neg: 1194 (%0.02:%0.19), avgs: 16 032: dt: 0.845, sse: 83109.5 (0.380, 31.6, 0.535, 0.344), neg: 1176 (%0.02:%0.19), avgs: 16 integrating with navgs=4 and tol=3.494e-02 033: dt: 0.700, sse: 83081.6 (0.381, 31.6, 0.536, 0.340), neg: 1147 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 034: dt: 0.517, sse: 83064.4 (0.382, 31.7, 0.536, 0.338), neg: 1204 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 035: dt: 0.041, sse: 83050.5 (0.382, 31.7, 0.537, 0.337), neg: 1173 (%0.02:%0.18), avgs: 0 036: dt: 0.042, sse: 83039.7 (0.382, 31.7, 0.537, 0.337), neg: 1171 (%0.02:%0.18), avgs: 0
blurring surfaces with sigma=1.00... done. curvature mean = 0.033, std = 0.955 curvature mean = 0.005, std = 0.964 tol=5.0e-01, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 037: dt: 115.451, sse: 87076.6 (0.383, 31.7, 0.537, 0.378), neg: 1086 (%0.02:%0.16), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 038: dt: 0.037, sse: 87076.2 (0.383, 31.7, 0.537, 0.378), neg: 1087 (%0.02:%0.16), avgs: 256 integrating with navgs=64 and tol=1.260e-01 039: dt: 6.000, sse: 87067.2 (0.385, 31.9, 0.540, 0.372), neg: 1155 (%0.02:%0.17), avgs: 64 integrating with navgs=16 and tol=6.442e-02 040: dt: 0.771, sse: 87047.6 (0.385, 32.0, 0.540, 0.370), neg: 1166 (%0.02:%0.17), avgs: 16 integrating with navgs=4 and tol=3.494e-02 041: dt: 0.926, sse: 86979.7 (0.387, 32.1, 0.542, 0.365), neg: 1169 (%0.02:%0.17), avgs: 4 042: dt: 0.550, sse: 86964.0 (0.388, 32.2, 0.543, 0.362), neg: 1182 (%0.02:%0.17), avgs: 4 integrating with navgs=1 and tol=2.210e-02 043: dt: 0.255, sse: 86957.1 (0.389, 32.2, 0.544, 0.360), neg: 1203 (%0.02:%0.18), avgs: 1 integrating with navgs=0 and tol=1.562e-02 044: dt: 0.035, sse: 86943.6 (0.389, 32.2, 0.544, 0.359), neg: 1219 (%0.02:%0.17), avgs: 0
blurring surfaces with sigma=0.50... done. curvature mean = 0.033, std = 0.965 curvature mean = 0.001, std = 0.983 tol=5.0e-01, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 045: dt: 39.869, sse: 89122.8 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 046: dt: 0.040, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 256 integrating with navgs=64 and tol=1.260e-01 047: dt: 0.015, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 64 integrating with navgs=16 and tol=6.442e-02 048: dt: 0.763, sse: 89114.0 (0.390, 32.3, 0.545, 0.379), neg: 1259 (%0.02:%0.18), avgs: 16 integrating with navgs=4 and tol=3.494e-02 049: dt: 0.531, sse: 89093.1 (0.391, 32.4, 0.546, 0.376), neg: 1286 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 050: dt: 0.280, sse: 89084.5 (0.392, 32.4, 0.547, 0.374), neg: 1315 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 051: dt: 0.128, sse: 89052.0 (0.393, 32.5, 0.547, 0.372), neg: 1300 (%0.04:%0.18), avgs: 0 052: dt: 0.032, sse: 89024.1 (0.392, 32.5, 0.547, 0.372), neg: 1329 (%0.02:%0.18), avgs: 0 053: dt: 0.033, sse: 89014.9 (0.393, 32.5, 0.547, 0.371), neg: 1377 (%0.02:%0.19), avgs: 0 2 Reading smoothwm curvature mean = -0.027, std = 0.269 calculating curvature of smoothwm surface tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization
blurring surfaces with sigma=4.00... done. curvature mean = 0.089, std = 0.338 curvature mean = 0.065, std = 0.378 tol=1.0e+00, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 054: dt: 3.299, sse: 745272.2 (0.384, 31.8, 0.538, 0.745), neg: 1168 (%0.04:%0.21), avgs: 1024 055: dt: 2.605, sse: 738110.4 (0.379, 31.4, 0.532, 0.748), neg: 740 (%0.01:%0.11), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 056: dt: 2.945, sse: 732954.1 (0.377, 31.2, 0.529, 0.751), neg: 774 (%0.02:%0.13), avgs: 256 057: dt: 2.657, sse: 728543.9 (0.374, 30.9, 0.526, 0.753), neg: 511 (%0.01:%0.08), avgs: 256 058: dt: 2.797, sse: 724908.4 (0.373, 30.8, 0.523, 0.756), neg: 551 (%0.01:%0.09), avgs: 256 059: dt: 2.740, sse: 721596.7 (0.371, 30.6, 0.521, 0.758), neg: 404 (%0.00:%0.06), avgs: 256 integrating with navgs=64 and tol=2.519e-01 060: dt: 0.737, sse: 720241.4 (0.370, 30.4, 0.520, 0.758), neg: 307 (%0.00:%0.05), avgs: 64 integrating with navgs=16 and tol=1.288e-01 061: dt: 6.542, sse: 716147.8 (0.370, 30.3, 0.518, 0.758), neg: 559 (%0.01:%0.07), avgs: 16 062: dt: 1.625, sse: 712606.1 (0.367, 29.9, 0.514, 0.757), neg: 151 (%0.00:%0.02), avgs: 16 063: dt: 11.631, sse: 707697.2 (0.368, 29.9, 0.513, 0.757), neg: 580 (%0.02:%0.06), avgs: 16 064: dt: 1.462, sse: 703515.8 (0.365, 29.3, 0.508, 0.756), neg: 88 (%0.00:%0.01), avgs: 16 065: dt: 29.856, sse: 694749.9 (0.368, 29.3, 0.505, 0.756), neg: 702 (%0.03:%0.07), avgs: 16 066: dt: 1.308, sse: 688816.4 (0.363, 28.4, 0.499, 0.756), neg: 56 (%0.00:%0.01), avgs: 16 067: dt: 25.204, sse: 684304.4 (0.364, 28.4, 0.496, 0.756), neg: 526 (%0.02:%0.06), avgs: 16 068: dt: 1.261, sse: 680584.3 (0.361, 27.8, 0.492, 0.756), neg: 60 (%0.00:%0.01), avgs: 16 069: dt: 10.120, sse: 679053.8 (0.362, 27.8, 0.492, 0.756), neg: 168 (%0.00:%0.01), avgs: 16 070: dt: 1.667, sse: 677616.7 (0.361, 27.6, 0.490, 0.756), neg: 51 (%0.00:%0.00), avgs: 16 071: dt: 6.724, sse: 676628.9 (0.361, 27.6, 0.490, 0.756), neg: 124 (%0.00:%0.01), avgs: 16 072: dt: 1.678, sse: 675656.8 (0.360, 27.5, 0.489, 0.756), neg: 52 (%0.00:%0.01), avgs: 16 073: dt: 8.216, sse: 674570.9 (0.361, 27.5, 0.488, 0.757), neg: 133 (%0.00:%0.01), avgs: 16 074: dt: 1.694, sse: 673512.5 (0.360, 27.3, 0.487, 0.757), neg: 53 (%0.00:%0.01), avgs: 16 075: dt: 1.764, sse: 673109.5 (0.360, 27.3, 0.487, 0.757), neg: 50 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 076: dt: 22.423, sse: 670854.9 (0.362, 27.2, 0.485, 0.742), neg: 200 (%0.00:%0.02), avgs: 4 077: dt: 1.661, sse: 669613.8 (0.361, 27.0, 0.484, 0.743), neg: 95 (%0.00:%0.01), avgs: 4 078: dt: 4.100, sse: 668786.9 (0.361, 27.0, 0.483, 0.744), neg: 92 (%0.00:%0.01), avgs: 4 079: dt: 3.000, sse: 668304.8 (0.361, 27.0, 0.483, 0.744), neg: 102 (%0.00:%0.01), avgs: 4 080: dt: 1.848, sse: 667791.1 (0.361, 26.9, 0.482, 0.744), neg: 77 (%0.00:%0.01), avgs: 4 081: dt: 5.400, sse: 667120.4 (0.361, 26.9, 0.482, 0.745), neg: 105 (%0.00:%0.01), avgs: 4 082: dt: 1.967, sse: 666614.0 (0.361, 26.8, 0.481, 0.745), neg: 73 (%0.00:%0.01), avgs: 4 083: dt: 4.462, sse: 666103.5 (0.361, 26.8, 0.481, 0.745), neg: 103 (%0.00:%0.01), avgs: 4 084: dt: 2.154, sse: 665621.2 (0.361, 26.8, 0.481, 0.745), neg: 76 (%0.00:%0.01), avgs: 4 085: dt: 2.615, sse: 665218.2 (0.361, 26.7, 0.480, 0.745), neg: 74 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 086: dt: 2.667, sse: 664951.9 (0.361, 26.7, 0.480, 0.744), neg: 76 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 087: dt: 0.506, sse: 664875.2 (0.361, 26.7, 0.480, 0.743), neg: 76 (%0.00:%0.01), avgs: 0
blurring surfaces with sigma=2.00... done. curvature mean = 0.090, std = 0.502 curvature mean = 0.020, std = 0.575 tol=1.0e+00, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 088: dt: 6.568, sse: 665480.4 (0.359, 26.7, 0.479, 0.937), neg: 270 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 089: dt: 1.744, sse: 664408.7 (0.359, 26.6, 0.478, 0.938), neg: 58 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 090: dt: 6.774, sse: 663527.8 (0.360, 26.6, 0.478, 0.941), neg: 186 (%0.01:%0.03), avgs: 64 integrating with navgs=16 and tol=1.288e-01 091: dt: 1.450, sse: 662806.6 (0.359, 26.5, 0.477, 0.939), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 092: dt: 6.720, sse: 662316.9 (0.359, 26.4, 0.477, 0.920), neg: 63 (%0.00:%0.01), avgs: 4 093: dt: 4.200, sse: 662053.0 (0.360, 26.5, 0.477, 0.913), neg: 85 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 094: dt: 0.903, sse: 661962.4 (0.360, 26.5, 0.477, 0.910), neg: 79 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 095: dt: 0.110, sse: 661952.2 (0.360, 26.5, 0.477, 0.909), neg: 73 (%0.00:%0.01), avgs: 0
blurring surfaces with sigma=1.00... done. curvature mean = 0.091, std = 0.611 curvature mean = 0.012, std = 0.719 tol=1.0e+00, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 096: dt: 8.627, sse: 661315.6 (0.359, 26.4, 0.476, 1.011), neg: 227 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 097: dt: 1.786, sse: 660256.5 (0.358, 26.3, 0.475, 1.012), neg: 66 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 098: dt: 5.077, sse: 659704.2 (0.359, 26.3, 0.474, 1.011), neg: 149 (%0.01:%0.02), avgs: 64 integrating with navgs=16 and tol=1.288e-01 099: dt: 1.431, sse: 659144.8 (0.358, 26.2, 0.474, 1.009), neg: 44 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 100: dt: 2.714, sse: 658944.9 (0.358, 26.2, 0.474, 0.996), neg: 47 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 101: dt: 0.075, sse: 658940.8 (0.358, 26.2, 0.474, 0.995), neg: 47 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 102: dt: 0.054, sse: 658935.4 (0.358, 26.2, 0.474, 0.994), neg: 47 (%0.00:%0.00), avgs: 0
blurring surfaces with sigma=0.50... done. curvature mean = 0.092, std = 0.682 curvature mean = 0.004, std = 0.828 tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 103: dt: 27.636, sse: 656331.2 (0.358, 26.1, 0.472, 1.092), neg: 165 (%0.01:%0.03), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 104: dt: 2.000, sse: 654661.2 (0.357, 25.9, 0.470, 1.083), neg: 82 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 105: dt: 3.105, sse: 653928.0 (0.357, 25.8, 0.469, 1.076), neg: 67 (%0.00:%0.01), avgs: 64 integrating with navgs=16 and tol=1.288e-01 106: dt: 2.292, sse: 653418.9 (0.357, 25.8, 0.469, 1.066), neg: 53 (%0.00:%0.01), avgs: 16 integrating with navgs=4 and tol=6.988e-02 107: dt: 1.047, sse: 653125.0 (0.357, 25.8, 0.469, 1.057), neg: 35 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 108: dt: 0.724, sse: 653027.9 (0.357, 25.8, 0.469, 1.047), neg: 36 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 109: dt: 0.064, sse: 653009.2 (0.357, 25.8, 0.469, 1.044), neg: 34 (%0.00:%0.00), avgs: 0
Removing remaining folds... tol=1.0e-01, sigma=0.5, host=athen, nav=64, nbrs=1, l_extern=10000.000, l_nlarea=100.000, l_corr=0.001, l_spring=0.005, l_dist=0.002 using quadratic fit line minimization nlarea/corr = 199999.984 integrating with navgs=64 and tol=2.519e-02 110: dt: 5.514, sse: 187681.1 (0.355, 25.8, 0.466, 1.062), neg: 85 (%0.00:%0.01), avgs: 64 111: dt: 1.735, sse: 184602.0 (0.355, 25.7, 0.466, 1.064), neg: 16 (%0.00:%0.00), avgs: 64 112: dt: 2.802, sse: 182518.2 (0.354, 25.7, 0.466, 1.068), neg: 16 (%0.00:%0.00), avgs: 64 113: dt: 3.973, sse: 180638.1 (0.355, 25.8, 0.466, 1.072), neg: 16 (%0.00:%0.00), avgs: 64 114: dt: 2.388, sse: 179649.5 (0.355, 25.8, 0.467, 1.075), neg: 20 (%0.00:%0.00), avgs: 64 115: dt: 3.088, sse: 178962.5 (0.355, 25.9, 0.467, 1.079), neg: 28 (%0.00:%0.00), avgs: 64 116: dt: 2.775, sse: 178633.9 (0.356, 26.0, 0.468, 1.082), neg: 36 (%0.00:%0.00), avgs: 64 117: dt: 0.831, sse: 178379.4 (0.356, 26.0, 0.468, 1.083), neg: 37 (%0.00:%0.00), avgs: 64 118: dt: 2.731, sse: 178093.2 (0.356, 26.1, 0.469, 1.087), neg: 52 (%0.00:%0.00), avgs: 64 119: dt: 0.606, sse: 177994.0 (0.356, 26.1, 0.469, 1.088), neg: 46 (%0.00:%0.00), avgs: 64 120: dt: 0.250, sse: 177987.2 (0.356, 26.1, 0.469, 1.088), neg: 48 (%0.00:%0.00), avgs: 64 integrating with navgs=16 and tol=1.288e-02 121: dt: 0.230, sse: 177851.4 (0.357, 26.1, 0.470, 1.089), neg: 46 (%0.00:%0.00), avgs: 16 122: dt: 0.031, sse: 177844.3 (0.357, 26.1, 0.470, 1.089), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-03 123: dt: 0.013, sse: 177801.5 (0.357, 26.1, 0.470, 1.089), neg: 38 (%0.00:%0.00), avgs: 4 124: dt: 0.014, sse: 177784.2 (0.357, 26.1, 0.470, 1.089), neg: 32 (%0.00:%0.00), avgs: 4 125: dt: 0.014, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-03 126: dt: 0.000, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-03 127: dt: 0.000, sse: 177746.4 (0.357, 26.1, 0.470, 1.089), neg: 30 (%0.00:%0.00), avgs: 0 128: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 129: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 registration took 0.25 hours MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25_parea0.reg... registration took 0.25 hours [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray%
-- 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 6, 2015, at 2:16 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
try doing:
setenv DIAG 0x04040
then run it again and send me the output
Bruce
On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote:
Dear Bruce, Thank again for your kind help. I’m running version 5.1.0. Below is the output of the command. I also ran it with parea=0 but almost no change in the result (see the attachment-1) I also plot the sluci map for comparison (see the attachment-2) but still not satisfactory.[IMAGE][IMAGE] Best ~/Data/P00001613/FreeSurferClean/surf> mris_register -1 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25.reg treating target as a single subject's surface... using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0 -------------------- tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 Starting MRISrigidBodyAlignGlobal() d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=0.9987 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.5086 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.0493 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.5976 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.1563 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=3.6949 MRISrigidBodyAlignGlobal() done 4.24 min curvature mean = 0.036, std = 0.923 curvature mean = 0.010, std = 0.931 curvature mean = 0.036, std = 0.951 curvature mean = 0.004, std = 0.967 curvature mean = 0.036, std = 0.962 curvature mean = 0.001, std = 0.984 2 Reading smoothwm curvature mean = -0.027, std = 0.269 tol=1.0e+00, sigma=0.5, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization curvature mean = 0.090, std = 0.337 curvature mean = 0.059, std = 0.386 curvature mean = 0.091, std = 0.507 curvature mean = 0.020, std = 0.580 curvature mean = 0.091, std = 0.618 curvature mean = 0.011, std = 0.723 curvature mean = 0.093, std = 0.681 curvature mean = 0.004, std = 0.828 MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25.reg... -- 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 4, 2015, at 7:30 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small)
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here?
Best
On Nov 4, 2015, at 5:56 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- 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.
_______________________________________________ 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.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ 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. _______________________________________________ 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.
did you try setting dist to .01 or something very small? Note that we don't register individual surfaces to each other for a reason - the inverse variance weighting in the warp functional is critical to it being stable and accurate
cheers Bruce
On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote:
Dear Bruce, I did set the DIAG and ran it again with no luck. Below is the output and attached please find the result.
Best[IMAGE]
ray@athens-b8-e8-56-47-e6-54:~> tcsh [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~] ray% setenv DIAG 0x04040 [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~/Data/P00001613/FreeSurferClean] ray% cd surf/ [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% mris_register -1 -parea 0 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25_parea0.reg treating target as a single subject's surface... using l_parea = 0.000 using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0
tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 reading precomputed curvature from lh.sulc
blurring surfaces with sigma=4.00... done. curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 finding optimal rigid alignment Starting MRISrigidBodyAlignGlobal() 000: dt: 0.000, sse: 135379.3 (0.279, 21.6, 0.412, 0.853), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 64.00 degree nbhd, min sse = 95165.92 (+64.00, +64.00, -64.00), min @ (0.00, 0.00, 0.00) = 95165.9 scanning 32.00 degree nbhd, min sse = 95165.92 (+32.00, +32.00, -32.00), min @ (8.00, 0.00, 0.00) = 90765.7 d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=1.1635 min sse = 90765.70 at (8.00, 0.00, 0.00) 001: dt: 0.000, sse: 130979.1 (0.279, 21.6, 0.412, 0.833), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 16.00 degree nbhd, min sse = 90765.70 (+16.00, +16.00, -16.00), min @ (-4.00, -4.00, 0.00) = 87371.6 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.7380 min sse = 87371.58 at (-4.00, -4.00, 0.00) 002: dt: 0.000, sse: 127584.9 (0.279, 21.6, 0.412, 0.817), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 8.00 degree nbhd, min sse = 87371.58 (+8.00, +8.00, -8.00), min @ (-2.00, 0.00, 0.00) = 86754.0 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.3264 min sse = 86754.03 at (-2.00, 0.00, 0.00) 003: dt: 0.000, sse: 126967.4 (0.279, 21.6, 0.412, 0.814), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 4.00 degree nbhd, min sse = 86754.03 (+4.00, +4.00, -4.00), min @ (1.00, 1.00, 0.00) = 86449.9 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.9298 min sse = 86449.91 at (1.00, 1.00, 0.00) 004: dt: 0.000, sse: 126663.3 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 2.00 degree nbhd, min sse = 86449.91 (+2.00, +2.00, -2.00), min @ (0.00, -0.50, 0.50) = 86436.5 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.5333 min sse = 86436.46 at (0.00, -0.50, 0.50) 005: dt: 0.000, sse: 126649.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 1.00 degree nbhd, min sse = 86436.45 (+1.00, +1.00, -1.00), min @ (0.00, 0.25, -0.25) = 86423.4 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=4.1242 min sse = 86423.41 at (0.00, 0.25, -0.25) 006: dt: 0.000, sse: 126636.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 0.50 degree nbhd, min sse = 86423.41 (+0.50, +0.50, -0.50), min @ (0.00, 0.00, 0.00) = 86423.4 MRISrigidBodyAlignGlobal() done 4.71 min tol=5.0e-01, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 007: dt: 241.975, sse: 109215.9 (0.308, 23.5, 0.429, 0.709), neg: 48 (%0.00:%0.01), avgs: 1024 008: dt: 117.533, sse: 99575.6 (0.298, 23.7, 0.432, 0.651), neg: 50 (%0.00:%0.01), avgs: 1024 009: dt: 136.508, sse: 95352.8 (0.305, 24.1, 0.436, 0.620), neg: 101 (%0.00:%0.02), avgs: 1024 010: dt: 105.985, sse: 92867.4 (0.303, 24.8, 0.442, 0.597), neg: 176 (%0.01:%0.03), avgs: 1024 011: dt: 149.343, sse: 90631.3 (0.311, 25.4, 0.448, 0.573), neg: 334 (%0.01:%0.06), avgs: 1024 012: dt: 85.796, sse: 89225.0 (0.312, 25.9, 0.453, 0.557), neg: 439 (%0.02:%0.08), avgs: 1024 013: dt: 177.745, sse: 87895.2 (0.321, 26.7, 0.460, 0.536), neg: 690 (%0.04:%0.13), avgs: 1024 014: dt: 66.370, sse: 87281.5 (0.322, 27.0, 0.463, 0.527), neg: 782 (%0.05:%0.14), avgs: 1024 015: dt: 236.440, sse: 86602.8 (0.331, 27.8, 0.472, 0.508), neg: 1155 (%0.08:%0.22), avgs: 1024 016: dt: 69.671, sse: 86419.9 (0.332, 28.0, 0.474, 0.502), neg: 1213 (%0.09:%0.23), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 017: dt: 122.246, sse: 82435.7 (0.347, 29.5, 0.492, 0.436), neg: 2026 (%0.16:%0.43), avgs: 256 018: dt: 69.015, sse: 81795.8 (0.350, 29.8, 0.499, 0.417), neg: 1971 (%0.12:%0.42), avgs: 256 019: dt: 36.896, sse: 81619.9 (0.353, 30.1, 0.502, 0.407), neg: 2059 (%0.12:%0.44), avgs: 256 integrating with navgs=64 and tol=1.260e-01 020: dt: 33.532, sse: 80387.5 (0.362, 30.7, 0.513, 0.370), neg: 2300 (%0.10:%0.49), avgs: 64 021: dt: 9.860, sse: 80325.9 (0.364, 30.9, 0.516, 0.362), neg: 2391 (%0.10:%0.51), avgs: 64 integrating with navgs=16 and tol=6.442e-02 022: dt: 6.761, sse: 80080.1 (0.368, 30.9, 0.520, 0.348), neg: 1935 (%0.07:%0.41), avgs: 16 023: dt: 0.929, sse: 80074.6 (0.368, 31.0, 0.521, 0.347), neg: 1928 (%0.06:%0.40), avgs: 16 integrating with navgs=4 and tol=3.494e-02 024: dt: 1.444, sse: 80034.9 (0.370, 31.0, 0.522, 0.343), neg: 1734 (%0.06:%0.35), avgs: 4 025: dt: 0.556, sse: 80028.6 (0.370, 31.0, 0.523, 0.341), neg: 1762 (%0.05:%0.36), avgs: 4 integrating with navgs=1 and tol=2.210e-02 026: dt: 0.066, sse: 80027.9 (0.370, 31.0, 0.523, 0.341), neg: 1753 (%0.05:%0.36), avgs: 1 integrating with navgs=0 and tol=1.562e-02 027: dt: 0.050, sse: 80016.6 (0.370, 31.0, 0.523, 0.341), neg: 1716 (%0.05:%0.33), avgs: 0
blurring surfaces with sigma=2.00... done. curvature mean = 0.035, std = 0.927 curvature mean = 0.011, std = 0.928 tol=5.0e-01, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 028: dt: 137.600, sse: 83361.7 (0.371, 31.1, 0.524, 0.373), neg: 1570 (%0.04:%0.30), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 029: dt: 16.658, sse: 83344.4 (0.372, 31.1, 0.526, 0.370), neg: 1484 (%0.03:%0.28), avgs: 256 integrating with navgs=64 and tol=1.260e-01 030: dt: 10.485, sse: 83245.6 (0.375, 31.4, 0.530, 0.359), neg: 1316 (%0.03:%0.23), avgs: 64 integrating with navgs=16 and tol=6.442e-02 031: dt: 4.712, sse: 83112.1 (0.379, 31.6, 0.534, 0.346), neg: 1194 (%0.02:%0.19), avgs: 16 032: dt: 0.845, sse: 83109.5 (0.380, 31.6, 0.535, 0.344), neg: 1176 (%0.02:%0.19), avgs: 16 integrating with navgs=4 and tol=3.494e-02 033: dt: 0.700, sse: 83081.6 (0.381, 31.6, 0.536, 0.340), neg: 1147 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 034: dt: 0.517, sse: 83064.4 (0.382, 31.7, 0.536, 0.338), neg: 1204 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 035: dt: 0.041, sse: 83050.5 (0.382, 31.7, 0.537, 0.337), neg: 1173 (%0.02:%0.18), avgs: 0 036: dt: 0.042, sse: 83039.7 (0.382, 31.7, 0.537, 0.337), neg: 1171 (%0.02:%0.18), avgs: 0
blurring surfaces with sigma=1.00... done. curvature mean = 0.033, std = 0.955 curvature mean = 0.005, std = 0.964 tol=5.0e-01, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 037: dt: 115.451, sse: 87076.6 (0.383, 31.7, 0.537, 0.378), neg: 1086 (%0.02:%0.16), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 038: dt: 0.037, sse: 87076.2 (0.383, 31.7, 0.537, 0.378), neg: 1087 (%0.02:%0.16), avgs: 256 integrating with navgs=64 and tol=1.260e-01 039: dt: 6.000, sse: 87067.2 (0.385, 31.9, 0.540, 0.372), neg: 1155 (%0.02:%0.17), avgs: 64 integrating with navgs=16 and tol=6.442e-02 040: dt: 0.771, sse: 87047.6 (0.385, 32.0, 0.540, 0.370), neg: 1166 (%0.02:%0.17), avgs: 16 integrating with navgs=4 and tol=3.494e-02 041: dt: 0.926, sse: 86979.7 (0.387, 32.1, 0.542, 0.365), neg: 1169 (%0.02:%0.17), avgs: 4 042: dt: 0.550, sse: 86964.0 (0.388, 32.2, 0.543, 0.362), neg: 1182 (%0.02:%0.17), avgs: 4 integrating with navgs=1 and tol=2.210e-02 043: dt: 0.255, sse: 86957.1 (0.389, 32.2, 0.544, 0.360), neg: 1203 (%0.02:%0.18), avgs: 1 integrating with navgs=0 and tol=1.562e-02 044: dt: 0.035, sse: 86943.6 (0.389, 32.2, 0.544, 0.359), neg: 1219 (%0.02:%0.17), avgs: 0
blurring surfaces with sigma=0.50... done. curvature mean = 0.033, std = 0.965 curvature mean = 0.001, std = 0.983 tol=5.0e-01, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 045: dt: 39.869, sse: 89122.8 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 046: dt: 0.040, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 256 integrating with navgs=64 and tol=1.260e-01 047: dt: 0.015, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 64 integrating with navgs=16 and tol=6.442e-02 048: dt: 0.763, sse: 89114.0 (0.390, 32.3, 0.545, 0.379), neg: 1259 (%0.02:%0.18), avgs: 16 integrating with navgs=4 and tol=3.494e-02 049: dt: 0.531, sse: 89093.1 (0.391, 32.4, 0.546, 0.376), neg: 1286 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 050: dt: 0.280, sse: 89084.5 (0.392, 32.4, 0.547, 0.374), neg: 1315 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 051: dt: 0.128, sse: 89052.0 (0.393, 32.5, 0.547, 0.372), neg: 1300 (%0.04:%0.18), avgs: 0 052: dt: 0.032, sse: 89024.1 (0.392, 32.5, 0.547, 0.372), neg: 1329 (%0.02:%0.18), avgs: 0 053: dt: 0.033, sse: 89014.9 (0.393, 32.5, 0.547, 0.371), neg: 1377 (%0.02:%0.19), avgs: 0 2 Reading smoothwm curvature mean = -0.027, std = 0.269 calculating curvature of smoothwm surface tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization
blurring surfaces with sigma=4.00... done. curvature mean = 0.089, std = 0.338 curvature mean = 0.065, std = 0.378 tol=1.0e+00, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 054: dt: 3.299, sse: 745272.2 (0.384, 31.8, 0.538, 0.745), neg: 1168 (%0.04:%0.21), avgs: 1024 055: dt: 2.605, sse: 738110.4 (0.379, 31.4, 0.532, 0.748), neg: 740 (%0.01:%0.11), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 056: dt: 2.945, sse: 732954.1 (0.377, 31.2, 0.529, 0.751), neg: 774 (%0.02:%0.13), avgs: 256 057: dt: 2.657, sse: 728543.9 (0.374, 30.9, 0.526, 0.753), neg: 511 (%0.01:%0.08), avgs: 256 058: dt: 2.797, sse: 724908.4 (0.373, 30.8, 0.523, 0.756), neg: 551 (%0.01:%0.09), avgs: 256 059: dt: 2.740, sse: 721596.7 (0.371, 30.6, 0.521, 0.758), neg: 404 (%0.00:%0.06), avgs: 256 integrating with navgs=64 and tol=2.519e-01 060: dt: 0.737, sse: 720241.4 (0.370, 30.4, 0.520, 0.758), neg: 307 (%0.00:%0.05), avgs: 64 integrating with navgs=16 and tol=1.288e-01 061: dt: 6.542, sse: 716147.8 (0.370, 30.3, 0.518, 0.758), neg: 559 (%0.01:%0.07), avgs: 16 062: dt: 1.625, sse: 712606.1 (0.367, 29.9, 0.514, 0.757), neg: 151 (%0.00:%0.02), avgs: 16 063: dt: 11.631, sse: 707697.2 (0.368, 29.9, 0.513, 0.757), neg: 580 (%0.02:%0.06), avgs: 16 064: dt: 1.462, sse: 703515.8 (0.365, 29.3, 0.508, 0.756), neg: 88 (%0.00:%0.01), avgs: 16 065: dt: 29.856, sse: 694749.9 (0.368, 29.3, 0.505, 0.756), neg: 702 (%0.03:%0.07), avgs: 16 066: dt: 1.308, sse: 688816.4 (0.363, 28.4, 0.499, 0.756), neg: 56 (%0.00:%0.01), avgs: 16 067: dt: 25.204, sse: 684304.4 (0.364, 28.4, 0.496, 0.756), neg: 526 (%0.02:%0.06), avgs: 16 068: dt: 1.261, sse: 680584.3 (0.361, 27.8, 0.492, 0.756), neg: 60 (%0.00:%0.01), avgs: 16 069: dt: 10.120, sse: 679053.8 (0.362, 27.8, 0.492, 0.756), neg: 168 (%0.00:%0.01), avgs: 16 070: dt: 1.667, sse: 677616.7 (0.361, 27.6, 0.490, 0.756), neg: 51 (%0.00:%0.00), avgs: 16 071: dt: 6.724, sse: 676628.9 (0.361, 27.6, 0.490, 0.756), neg: 124 (%0.00:%0.01), avgs: 16 072: dt: 1.678, sse: 675656.8 (0.360, 27.5, 0.489, 0.756), neg: 52 (%0.00:%0.01), avgs: 16 073: dt: 8.216, sse: 674570.9 (0.361, 27.5, 0.488, 0.757), neg: 133 (%0.00:%0.01), avgs: 16 074: dt: 1.694, sse: 673512.5 (0.360, 27.3, 0.487, 0.757), neg: 53 (%0.00:%0.01), avgs: 16 075: dt: 1.764, sse: 673109.5 (0.360, 27.3, 0.487, 0.757), neg: 50 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 076: dt: 22.423, sse: 670854.9 (0.362, 27.2, 0.485, 0.742), neg: 200 (%0.00:%0.02), avgs: 4 077: dt: 1.661, sse: 669613.8 (0.361, 27.0, 0.484, 0.743), neg: 95 (%0.00:%0.01), avgs: 4 078: dt: 4.100, sse: 668786.9 (0.361, 27.0, 0.483, 0.744), neg: 92 (%0.00:%0.01), avgs: 4 079: dt: 3.000, sse: 668304.8 (0.361, 27.0, 0.483, 0.744), neg: 102 (%0.00:%0.01), avgs: 4 080: dt: 1.848, sse: 667791.1 (0.361, 26.9, 0.482, 0.744), neg: 77 (%0.00:%0.01), avgs: 4 081: dt: 5.400, sse: 667120.4 (0.361, 26.9, 0.482, 0.745), neg: 105 (%0.00:%0.01), avgs: 4 082: dt: 1.967, sse: 666614.0 (0.361, 26.8, 0.481, 0.745), neg: 73 (%0.00:%0.01), avgs: 4 083: dt: 4.462, sse: 666103.5 (0.361, 26.8, 0.481, 0.745), neg: 103 (%0.00:%0.01), avgs: 4 084: dt: 2.154, sse: 665621.2 (0.361, 26.8, 0.481, 0.745), neg: 76 (%0.00:%0.01), avgs: 4 085: dt: 2.615, sse: 665218.2 (0.361, 26.7, 0.480, 0.745), neg: 74 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 086: dt: 2.667, sse: 664951.9 (0.361, 26.7, 0.480, 0.744), neg: 76 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 087: dt: 0.506, sse: 664875.2 (0.361, 26.7, 0.480, 0.743), neg: 76 (%0.00:%0.01), avgs: 0
blurring surfaces with sigma=2.00... done. curvature mean = 0.090, std = 0.502 curvature mean = 0.020, std = 0.575 tol=1.0e+00, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 088: dt: 6.568, sse: 665480.4 (0.359, 26.7, 0.479, 0.937), neg: 270 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 089: dt: 1.744, sse: 664408.7 (0.359, 26.6, 0.478, 0.938), neg: 58 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 090: dt: 6.774, sse: 663527.8 (0.360, 26.6, 0.478, 0.941), neg: 186 (%0.01:%0.03), avgs: 64 integrating with navgs=16 and tol=1.288e-01 091: dt: 1.450, sse: 662806.6 (0.359, 26.5, 0.477, 0.939), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 092: dt: 6.720, sse: 662316.9 (0.359, 26.4, 0.477, 0.920), neg: 63 (%0.00:%0.01), avgs: 4 093: dt: 4.200, sse: 662053.0 (0.360, 26.5, 0.477, 0.913), neg: 85 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 094: dt: 0.903, sse: 661962.4 (0.360, 26.5, 0.477, 0.910), neg: 79 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 095: dt: 0.110, sse: 661952.2 (0.360, 26.5, 0.477, 0.909), neg: 73 (%0.00:%0.01), avgs: 0
blurring surfaces with sigma=1.00... done. curvature mean = 0.091, std = 0.611 curvature mean = 0.012, std = 0.719 tol=1.0e+00, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 096: dt: 8.627, sse: 661315.6 (0.359, 26.4, 0.476, 1.011), neg: 227 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 097: dt: 1.786, sse: 660256.5 (0.358, 26.3, 0.475, 1.012), neg: 66 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 098: dt: 5.077, sse: 659704.2 (0.359, 26.3, 0.474, 1.011), neg: 149 (%0.01:%0.02), avgs: 64 integrating with navgs=16 and tol=1.288e-01 099: dt: 1.431, sse: 659144.8 (0.358, 26.2, 0.474, 1.009), neg: 44 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 100: dt: 2.714, sse: 658944.9 (0.358, 26.2, 0.474, 0.996), neg: 47 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 101: dt: 0.075, sse: 658940.8 (0.358, 26.2, 0.474, 0.995), neg: 47 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 102: dt: 0.054, sse: 658935.4 (0.358, 26.2, 0.474, 0.994), neg: 47 (%0.00:%0.00), avgs: 0
blurring surfaces with sigma=0.50... done. curvature mean = 0.092, std = 0.682 curvature mean = 0.004, std = 0.828 tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 103: dt: 27.636, sse: 656331.2 (0.358, 26.1, 0.472, 1.092), neg: 165 (%0.01:%0.03), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 104: dt: 2.000, sse: 654661.2 (0.357, 25.9, 0.470, 1.083), neg: 82 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 105: dt: 3.105, sse: 653928.0 (0.357, 25.8, 0.469, 1.076), neg: 67 (%0.00:%0.01), avgs: 64 integrating with navgs=16 and tol=1.288e-01 106: dt: 2.292, sse: 653418.9 (0.357, 25.8, 0.469, 1.066), neg: 53 (%0.00:%0.01), avgs: 16 integrating with navgs=4 and tol=6.988e-02 107: dt: 1.047, sse: 653125.0 (0.357, 25.8, 0.469, 1.057), neg: 35 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 108: dt: 0.724, sse: 653027.9 (0.357, 25.8, 0.469, 1.047), neg: 36 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 109: dt: 0.064, sse: 653009.2 (0.357, 25.8, 0.469, 1.044), neg: 34 (%0.00:%0.00), avgs: 0
Removing remaining folds... tol=1.0e-01, sigma=0.5, host=athen, nav=64, nbrs=1, l_extern=10000.000, l_nlarea=100.000, l_corr=0.001, l_spring=0.005, l_dist=0.002 using quadratic fit line minimization nlarea/corr = 199999.984 integrating with navgs=64 and tol=2.519e-02 110: dt: 5.514, sse: 187681.1 (0.355, 25.8, 0.466, 1.062), neg: 85 (%0.00:%0.01), avgs: 64 111: dt: 1.735, sse: 184602.0 (0.355, 25.7, 0.466, 1.064), neg: 16 (%0.00:%0.00), avgs: 64 112: dt: 2.802, sse: 182518.2 (0.354, 25.7, 0.466, 1.068), neg: 16 (%0.00:%0.00), avgs: 64 113: dt: 3.973, sse: 180638.1 (0.355, 25.8, 0.466, 1.072), neg: 16 (%0.00:%0.00), avgs: 64 114: dt: 2.388, sse: 179649.5 (0.355, 25.8, 0.467, 1.075), neg: 20 (%0.00:%0.00), avgs: 64 115: dt: 3.088, sse: 178962.5 (0.355, 25.9, 0.467, 1.079), neg: 28 (%0.00:%0.00), avgs: 64 116: dt: 2.775, sse: 178633.9 (0.356, 26.0, 0.468, 1.082), neg: 36 (%0.00:%0.00), avgs: 64 117: dt: 0.831, sse: 178379.4 (0.356, 26.0, 0.468, 1.083), neg: 37 (%0.00:%0.00), avgs: 64 118: dt: 2.731, sse: 178093.2 (0.356, 26.1, 0.469, 1.087), neg: 52 (%0.00:%0.00), avgs: 64 119: dt: 0.606, sse: 177994.0 (0.356, 26.1, 0.469, 1.088), neg: 46 (%0.00:%0.00), avgs: 64 120: dt: 0.250, sse: 177987.2 (0.356, 26.1, 0.469, 1.088), neg: 48 (%0.00:%0.00), avgs: 64 integrating with navgs=16 and tol=1.288e-02 121: dt: 0.230, sse: 177851.4 (0.357, 26.1, 0.470, 1.089), neg: 46 (%0.00:%0.00), avgs: 16 122: dt: 0.031, sse: 177844.3 (0.357, 26.1, 0.470, 1.089), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-03 123: dt: 0.013, sse: 177801.5 (0.357, 26.1, 0.470, 1.089), neg: 38 (%0.00:%0.00), avgs: 4 124: dt: 0.014, sse: 177784.2 (0.357, 26.1, 0.470, 1.089), neg: 32 (%0.00:%0.00), avgs: 4 125: dt: 0.014, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-03 126: dt: 0.000, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-03 127: dt: 0.000, sse: 177746.4 (0.357, 26.1, 0.470, 1.089), neg: 30 (%0.00:%0.00), avgs: 0 128: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 129: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 registration took 0.25 hours MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25_parea0.reg... registration took 0.25 hours [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray%
-- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University
Alt: razlighi@gmail.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/
On Nov 6, 2015, at 2:16 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Ray try doing: setenv DIAG 0x04040 then run it again and send me the output Bruce On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote: Dear Bruce, Thank again for your kind help. I’m running version 5.1.0. Below is the output of the command. I also ran it with parea=0 but almost no change in the result (see the attachment-1) I also plot the sluci map for comparison (see the attachment-2) but still not satisfactory.[IMAGE][IMAGE] Best ~/Data/P00001613/FreeSurferClean/surf\> mris_register -1 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25.reg treating target as a single subject's surface... using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0 -------------------- tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 Starting MRISrigidBodyAlignGlobal() d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=0.9987 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.5086 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.0493 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.5976 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.1563 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=3.6949 MRISrigidBodyAlignGlobal() done 4.24 min curvature mean = 0.036, std = 0.923 curvature mean = 0.010, std = 0.931 curvature mean = 0.036, std = 0.951 curvature mean = 0.004, std = 0.967 curvature mean = 0.036, std = 0.962 curvature mean = 0.001, std = 0.984 2 Reading smoothwm curvature mean = -0.027, std = 0.269 tol=1.0e+00, sigma=0.5, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization curvature mean = 0.090, std = 0.337 curvature mean = 0.059, std = 0.386 curvature mean = 0.091, std = 0.507 curvature mean = 0.020, std = 0.580 curvature mean = 0.091, std = 0.618 curvature mean = 0.011, std = 0.723 curvature mean = 0.093, std = 0.681 curvature mean = 0.004, std = 0.828 MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25.reg... -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: razlighi@gmail.com Office Phone: 212-342-1352 Office Fax: 212-342-1838 Website: http://www.columbia.edu/cu/qnl/ On Nov 4, 2015, at 7:30 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote: Hi Ray what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small) cheers Bruce On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote: Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here? Best On Nov 4, 2015, at 5:56 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote: Hi Ray that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps. Which inaccuracy are you referring to? cheers Bruce On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote: 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 [IMAGE] -- Ray Razlighi, Ph.D. Assistant Professor Quantitative Neuroimaging Laboratory Division of Cognitive Neuroscience Department of Neurology Columbia University Alt: 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.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.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.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinf o/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.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurf er
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 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 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.
Dear Bruce, The one I sent you were with diet .25. I reran it with .01 and now I can see minor non-linear shift ( see the attachment). I know that this registration is set up for the template one and I’m just trying to fins out the base parameter set for the subject to subject surface base registration. Since this is working perfectly for template base I assume by setting the right parameters I can get to something similar. The problem is that the curvature maps after registrations are so different that makes me believe that I’m still doing something wrong. I’m more than happy to try any other suggestion to fix this.
Best[cid:262429E6-1360-4819-B834-884EF40F404E@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 6, 2015, at 3:23 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
did you try setting dist to .01 or something very small? Note that we don't register individual surfaces to each other for a reason - the inverse variance weighting in the warp functional is critical to it being stable and accurate
cheers Bruce
On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote:
Dear Bruce, I did set the DIAG and ran it again with no luck. Below is the output and attached please find the result. Best[IMAGE] ray@athens-b8-e8-56-47-e6-54:~> tcsh [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~] ray% setenv DIAG 0x04040 [athens-b8-e8-56-47-e6-54:~] ray% [athens-b8-e8-56-47-e6-54:~/Data/P00001613/FreeSurferClean] ray% cd surf/ [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray% mris_register -1 -parea 0 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25_parea0.reg treating target as a single subject's surface... using l_parea = 0.000 using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0 -------------------- tol=5.0e-01, sigma=0.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 reading precomputed curvature from lh.sulc blurring surfaces with sigma=4.00... done. curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 finding optimal rigid alignment Starting MRISrigidBodyAlignGlobal() 000: dt: 0.000, sse: 135379.3 (0.279, 21.6, 0.412, 0.853), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 64.00 degree nbhd, min sse = 95165.92 (+64.00, +64.00, -64.00), min @ (0.00, 0.00, 0.00) = 95165.9 scanning 32.00 degree nbhd, min sse = 95165.92 (+32.00, +32.00, -32.00), min @ (8.00, 0.00, 0.00) = 90765.7 d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=1.1635 min sse = 90765.70 at (8.00, 0.00, 0.00) 001: dt: 0.000, sse: 130979.1 (0.279, 21.6, 0.412, 0.833), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 16.00 degree nbhd, min sse = 90765.70 (+16.00, +16.00, -16.00), min @ (-4.00, -4.00, 0.00) = 87371.6 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.7380 min sse = 87371.58 at (-4.00, -4.00, 0.00) 002: dt: 0.000, sse: 127584.9 (0.279, 21.6, 0.412, 0.817), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 8.00 degree nbhd, min sse = 87371.58 (+8.00, +8.00, -8.00), min @ (-2.00, 0.00, 0.00) = 86754.0 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.3264 min sse = 86754.03 at (-2.00, 0.00, 0.00) 003: dt: 0.000, sse: 126967.4 (0.279, 21.6, 0.412, 0.814), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 4.00 degree nbhd, min sse = 86754.03 (+4.00, +4.00, -4.00), min @ (1.00, 1.00, 0.00) = 86449.9 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.9298 min sse = 86449.91 at (1.00, 1.00, 0.00) 004: dt: 0.000, sse: 126663.3 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 2.00 degree nbhd, min sse = 86449.91 (+2.00, +2.00, -2.00), min @ (0.00, -0.50, 0.50) = 86436.5 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.5333 min sse = 86436.46 at (0.00, -0.50, 0.50) 005: dt: 0.000, sse: 126649.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 1.00 degree nbhd, min sse = 86436.45 (+1.00, +1.00, -1.00), min @ (0.00, 0.25, -0.25) = 86423.4 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=4.1242 min sse = 86423.41 at (0.00, 0.25, -0.25) 006: dt: 0.000, sse: 126636.8 (0.279, 21.6, 0.412, 0.813), neg: 0 (%0.00:%0.00), avgs: 1024 scanning 0.50 degree nbhd, min sse = 86423.41 (+0.50, +0.50, -0.50), min @ (0.00, 0.00, 0.00) = 86423.4 MRISrigidBodyAlignGlobal() done 4.71 min tol=5.0e-01, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 007: dt: 241.975, sse: 109215.9 (0.308, 23.5, 0.429, 0.709), neg: 48 (%0.00:%0.01), avgs: 1024 008: dt: 117.533, sse: 99575.6 (0.298, 23.7, 0.432, 0.651), neg: 50 (%0.00:%0.01), avgs: 1024 009: dt: 136.508, sse: 95352.8 (0.305, 24.1, 0.436, 0.620), neg: 101 (%0.00:%0.02), avgs: 1024 010: dt: 105.985, sse: 92867.4 (0.303, 24.8, 0.442, 0.597), neg: 176 (%0.01:%0.03), avgs: 1024 011: dt: 149.343, sse: 90631.3 (0.311, 25.4, 0.448, 0.573), neg: 334 (%0.01:%0.06), avgs: 1024 012: dt: 85.796, sse: 89225.0 (0.312, 25.9, 0.453, 0.557), neg: 439 (%0.02:%0.08), avgs: 1024 013: dt: 177.745, sse: 87895.2 (0.321, 26.7, 0.460, 0.536), neg: 690 (%0.04:%0.13), avgs: 1024 014: dt: 66.370, sse: 87281.5 (0.322, 27.0, 0.463, 0.527), neg: 782 (%0.05:%0.14), avgs: 1024 015: dt: 236.440, sse: 86602.8 (0.331, 27.8, 0.472, 0.508), neg: 1155 (%0.08:%0.22), avgs: 1024 016: dt: 69.671, sse: 86419.9 (0.332, 28.0, 0.474, 0.502), neg: 1213 (%0.09:%0.23), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 017: dt: 122.246, sse: 82435.7 (0.347, 29.5, 0.492, 0.436), neg: 2026 (%0.16:%0.43), avgs: 256 018: dt: 69.015, sse: 81795.8 (0.350, 29.8, 0.499, 0.417), neg: 1971 (%0.12:%0.42), avgs: 256 019: dt: 36.896, sse: 81619.9 (0.353, 30.1, 0.502, 0.407), neg: 2059 (%0.12:%0.44), avgs: 256 integrating with navgs=64 and tol=1.260e-01 020: dt: 33.532, sse: 80387.5 (0.362, 30.7, 0.513, 0.370), neg: 2300 (%0.10:%0.49), avgs: 64 021: dt: 9.860, sse: 80325.9 (0.364, 30.9, 0.516, 0.362), neg: 2391 (%0.10:%0.51), avgs: 64 integrating with navgs=16 and tol=6.442e-02 022: dt: 6.761, sse: 80080.1 (0.368, 30.9, 0.520, 0.348), neg: 1935 (%0.07:%0.41), avgs: 16 023: dt: 0.929, sse: 80074.6 (0.368, 31.0, 0.521, 0.347), neg: 1928 (%0.06:%0.40), avgs: 16 integrating with navgs=4 and tol=3.494e-02 024: dt: 1.444, sse: 80034.9 (0.370, 31.0, 0.522, 0.343), neg: 1734 (%0.06:%0.35), avgs: 4 025: dt: 0.556, sse: 80028.6 (0.370, 31.0, 0.523, 0.341), neg: 1762 (%0.05:%0.36), avgs: 4 integrating with navgs=1 and tol=2.210e-02 026: dt: 0.066, sse: 80027.9 (0.370, 31.0, 0.523, 0.341), neg: 1753 (%0.05:%0.36), avgs: 1 integrating with navgs=0 and tol=1.562e-02 027: dt: 0.050, sse: 80016.6 (0.370, 31.0, 0.523, 0.341), neg: 1716 (%0.05:%0.33), avgs: 0 blurring surfaces with sigma=2.00... done. curvature mean = 0.035, std = 0.927 curvature mean = 0.011, std = 0.928 tol=5.0e-01, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 028: dt: 137.600, sse: 83361.7 (0.371, 31.1, 0.524, 0.373), neg: 1570 (%0.04:%0.30), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 029: dt: 16.658, sse: 83344.4 (0.372, 31.1, 0.526, 0.370), neg: 1484 (%0.03:%0.28), avgs: 256 integrating with navgs=64 and tol=1.260e-01 030: dt: 10.485, sse: 83245.6 (0.375, 31.4, 0.530, 0.359), neg: 1316 (%0.03:%0.23), avgs: 64 integrating with navgs=16 and tol=6.442e-02 031: dt: 4.712, sse: 83112.1 (0.379, 31.6, 0.534, 0.346), neg: 1194 (%0.02:%0.19), avgs: 16 032: dt: 0.845, sse: 83109.5 (0.380, 31.6, 0.535, 0.344), neg: 1176 (%0.02:%0.19), avgs: 16 integrating with navgs=4 and tol=3.494e-02 033: dt: 0.700, sse: 83081.6 (0.381, 31.6, 0.536, 0.340), neg: 1147 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 034: dt: 0.517, sse: 83064.4 (0.382, 31.7, 0.536, 0.338), neg: 1204 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 035: dt: 0.041, sse: 83050.5 (0.382, 31.7, 0.537, 0.337), neg: 1173 (%0.02:%0.18), avgs: 0 036: dt: 0.042, sse: 83039.7 (0.382, 31.7, 0.537, 0.337), neg: 1171 (%0.02:%0.18), avgs: 0 blurring surfaces with sigma=1.00... done. curvature mean = 0.033, std = 0.955 curvature mean = 0.005, std = 0.964 tol=5.0e-01, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 037: dt: 115.451, sse: 87076.6 (0.383, 31.7, 0.537, 0.378), neg: 1086 (%0.02:%0.16), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 038: dt: 0.037, sse: 87076.2 (0.383, 31.7, 0.537, 0.378), neg: 1087 (%0.02:%0.16), avgs: 256 integrating with navgs=64 and tol=1.260e-01 039: dt: 6.000, sse: 87067.2 (0.385, 31.9, 0.540, 0.372), neg: 1155 (%0.02:%0.17), avgs: 64 integrating with navgs=16 and tol=6.442e-02 040: dt: 0.771, sse: 87047.6 (0.385, 32.0, 0.540, 0.370), neg: 1166 (%0.02:%0.17), avgs: 16 integrating with navgs=4 and tol=3.494e-02 041: dt: 0.926, sse: 86979.7 (0.387, 32.1, 0.542, 0.365), neg: 1169 (%0.02:%0.17), avgs: 4 042: dt: 0.550, sse: 86964.0 (0.388, 32.2, 0.543, 0.362), neg: 1182 (%0.02:%0.17), avgs: 4 integrating with navgs=1 and tol=2.210e-02 043: dt: 0.255, sse: 86957.1 (0.389, 32.2, 0.544, 0.360), neg: 1203 (%0.02:%0.18), avgs: 1 integrating with navgs=0 and tol=1.562e-02 044: dt: 0.035, sse: 86943.6 (0.389, 32.2, 0.544, 0.359), neg: 1219 (%0.02:%0.17), avgs: 0 blurring surfaces with sigma=0.50... done. curvature mean = 0.033, std = 0.965 curvature mean = 0.001, std = 0.983 tol=5.0e-01, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 1.000 integrating with navgs=1024 and tol=5.002e-01 045: dt: 39.869, sse: 89122.8 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 1024 integrating with navgs=256 and tol=2.505e-01 046: dt: 0.040, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 256 integrating with navgs=64 and tol=1.260e-01 047: dt: 0.015, sse: 89122.7 (0.389, 32.2, 0.544, 0.381), neg: 1239 (%0.02:%0.18), avgs: 64 integrating with navgs=16 and tol=6.442e-02 048: dt: 0.763, sse: 89114.0 (0.390, 32.3, 0.545, 0.379), neg: 1259 (%0.02:%0.18), avgs: 16 integrating with navgs=4 and tol=3.494e-02 049: dt: 0.531, sse: 89093.1 (0.391, 32.4, 0.546, 0.376), neg: 1286 (%0.02:%0.18), avgs: 4 integrating with navgs=1 and tol=2.210e-02 050: dt: 0.280, sse: 89084.5 (0.392, 32.4, 0.547, 0.374), neg: 1315 (%0.02:%0.19), avgs: 1 integrating with navgs=0 and tol=1.562e-02 051: dt: 0.128, sse: 89052.0 (0.393, 32.5, 0.547, 0.372), neg: 1300 (%0.04:%0.18), avgs: 0 052: dt: 0.032, sse: 89024.1 (0.392, 32.5, 0.547, 0.372), neg: 1329 (%0.02:%0.18), avgs: 0 053: dt: 0.033, sse: 89014.9 (0.393, 32.5, 0.547, 0.371), neg: 1377 (%0.02:%0.19), avgs: 0 2 Reading smoothwm curvature mean = -0.027, std = 0.269 calculating curvature of smoothwm surface tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization blurring surfaces with sigma=4.00... done. curvature mean = 0.089, std = 0.338 curvature mean = 0.065, std = 0.378 tol=1.0e+00, sigma=4.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 054: dt: 3.299, sse: 745272.2 (0.384, 31.8, 0.538, 0.745), neg: 1168 (%0.04:%0.21), avgs: 1024 055: dt: 2.605, sse: 738110.4 (0.379, 31.4, 0.532, 0.748), neg: 740 (%0.01:%0.11), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 056: dt: 2.945, sse: 732954.1 (0.377, 31.2, 0.529, 0.751), neg: 774 (%0.02:%0.13), avgs: 256 057: dt: 2.657, sse: 728543.9 (0.374, 30.9, 0.526, 0.753), neg: 511 (%0.01:%0.08), avgs: 256 058: dt: 2.797, sse: 724908.4 (0.373, 30.8, 0.523, 0.756), neg: 551 (%0.01:%0.09), avgs: 256 059: dt: 2.740, sse: 721596.7 (0.371, 30.6, 0.521, 0.758), neg: 404 (%0.00:%0.06), avgs: 256 integrating with navgs=64 and tol=2.519e-01 060: dt: 0.737, sse: 720241.4 (0.370, 30.4, 0.520, 0.758), neg: 307 (%0.00:%0.05), avgs: 64 integrating with navgs=16 and tol=1.288e-01 061: dt: 6.542, sse: 716147.8 (0.370, 30.3, 0.518, 0.758), neg: 559 (%0.01:%0.07), avgs: 16 062: dt: 1.625, sse: 712606.1 (0.367, 29.9, 0.514, 0.757), neg: 151 (%0.00:%0.02), avgs: 16 063: dt: 11.631, sse: 707697.2 (0.368, 29.9, 0.513, 0.757), neg: 580 (%0.02:%0.06), avgs: 16 064: dt: 1.462, sse: 703515.8 (0.365, 29.3, 0.508, 0.756), neg: 88 (%0.00:%0.01), avgs: 16 065: dt: 29.856, sse: 694749.9 (0.368, 29.3, 0.505, 0.756), neg: 702 (%0.03:%0.07), avgs: 16 066: dt: 1.308, sse: 688816.4 (0.363, 28.4, 0.499, 0.756), neg: 56 (%0.00:%0.01), avgs: 16 067: dt: 25.204, sse: 684304.4 (0.364, 28.4, 0.496, 0.756), neg: 526 (%0.02:%0.06), avgs: 16 068: dt: 1.261, sse: 680584.3 (0.361, 27.8, 0.492, 0.756), neg: 60 (%0.00:%0.01), avgs: 16 069: dt: 10.120, sse: 679053.8 (0.362, 27.8, 0.492, 0.756), neg: 168 (%0.00:%0.01), avgs: 16 070: dt: 1.667, sse: 677616.7 (0.361, 27.6, 0.490, 0.756), neg: 51 (%0.00:%0.00), avgs: 16 071: dt: 6.724, sse: 676628.9 (0.361, 27.6, 0.490, 0.756), neg: 124 (%0.00:%0.01), avgs: 16 072: dt: 1.678, sse: 675656.8 (0.360, 27.5, 0.489, 0.756), neg: 52 (%0.00:%0.01), avgs: 16 073: dt: 8.216, sse: 674570.9 (0.361, 27.5, 0.488, 0.757), neg: 133 (%0.00:%0.01), avgs: 16 074: dt: 1.694, sse: 673512.5 (0.360, 27.3, 0.487, 0.757), neg: 53 (%0.00:%0.01), avgs: 16 075: dt: 1.764, sse: 673109.5 (0.360, 27.3, 0.487, 0.757), neg: 50 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 076: dt: 22.423, sse: 670854.9 (0.362, 27.2, 0.485, 0.742), neg: 200 (%0.00:%0.02), avgs: 4 077: dt: 1.661, sse: 669613.8 (0.361, 27.0, 0.484, 0.743), neg: 95 (%0.00:%0.01), avgs: 4 078: dt: 4.100, sse: 668786.9 (0.361, 27.0, 0.483, 0.744), neg: 92 (%0.00:%0.01), avgs: 4 079: dt: 3.000, sse: 668304.8 (0.361, 27.0, 0.483, 0.744), neg: 102 (%0.00:%0.01), avgs: 4 080: dt: 1.848, sse: 667791.1 (0.361, 26.9, 0.482, 0.744), neg: 77 (%0.00:%0.01), avgs: 4 081: dt: 5.400, sse: 667120.4 (0.361, 26.9, 0.482, 0.745), neg: 105 (%0.00:%0.01), avgs: 4 082: dt: 1.967, sse: 666614.0 (0.361, 26.8, 0.481, 0.745), neg: 73 (%0.00:%0.01), avgs: 4 083: dt: 4.462, sse: 666103.5 (0.361, 26.8, 0.481, 0.745), neg: 103 (%0.00:%0.01), avgs: 4 084: dt: 2.154, sse: 665621.2 (0.361, 26.8, 0.481, 0.745), neg: 76 (%0.00:%0.01), avgs: 4 085: dt: 2.615, sse: 665218.2 (0.361, 26.7, 0.480, 0.745), neg: 74 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 086: dt: 2.667, sse: 664951.9 (0.361, 26.7, 0.480, 0.744), neg: 76 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 087: dt: 0.506, sse: 664875.2 (0.361, 26.7, 0.480, 0.743), neg: 76 (%0.00:%0.01), avgs: 0 blurring surfaces with sigma=2.00... done. curvature mean = 0.090, std = 0.502 curvature mean = 0.020, std = 0.575 tol=1.0e+00, sigma=2.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 088: dt: 6.568, sse: 665480.4 (0.359, 26.7, 0.479, 0.937), neg: 270 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 089: dt: 1.744, sse: 664408.7 (0.359, 26.6, 0.478, 0.938), neg: 58 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 090: dt: 6.774, sse: 663527.8 (0.360, 26.6, 0.478, 0.941), neg: 186 (%0.01:%0.03), avgs: 64 integrating with navgs=16 and tol=1.288e-01 091: dt: 1.450, sse: 662806.6 (0.359, 26.5, 0.477, 0.939), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 092: dt: 6.720, sse: 662316.9 (0.359, 26.4, 0.477, 0.920), neg: 63 (%0.00:%0.01), avgs: 4 093: dt: 4.200, sse: 662053.0 (0.360, 26.5, 0.477, 0.913), neg: 85 (%0.00:%0.01), avgs: 4 integrating with navgs=1 and tol=4.419e-02 094: dt: 0.903, sse: 661962.4 (0.360, 26.5, 0.477, 0.910), neg: 79 (%0.00:%0.01), avgs: 1 integrating with navgs=0 and tol=3.125e-02 095: dt: 0.110, sse: 661952.2 (0.360, 26.5, 0.477, 0.909), neg: 73 (%0.00:%0.01), avgs: 0 blurring surfaces with sigma=1.00... done. curvature mean = 0.091, std = 0.611 curvature mean = 0.012, std = 0.719 tol=1.0e+00, sigma=1.0, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 096: dt: 8.627, sse: 661315.6 (0.359, 26.4, 0.476, 1.011), neg: 227 (%0.02:%0.04), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 097: dt: 1.786, sse: 660256.5 (0.358, 26.3, 0.475, 1.012), neg: 66 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 098: dt: 5.077, sse: 659704.2 (0.359, 26.3, 0.474, 1.011), neg: 149 (%0.01:%0.02), avgs: 64 integrating with navgs=16 and tol=1.288e-01 099: dt: 1.431, sse: 659144.8 (0.358, 26.2, 0.474, 1.009), neg: 44 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-02 100: dt: 2.714, sse: 658944.9 (0.358, 26.2, 0.474, 0.996), neg: 47 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 101: dt: 0.075, sse: 658940.8 (0.358, 26.2, 0.474, 0.995), neg: 47 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 102: dt: 0.054, sse: 658935.4 (0.358, 26.2, 0.474, 0.994), neg: 47 (%0.00:%0.00), avgs: 0 blurring surfaces with sigma=0.50... done. curvature mean = 0.092, std = 0.682 curvature mean = 0.004, std = 0.828 tol=1.0e+00, sigma=0.5, host=athen, nav=1024, nbrs=1, l_extern=10000.000, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization nlarea/corr = 20.000 integrating with navgs=1024 and tol=1.000e+00 103: dt: 27.636, sse: 656331.2 (0.358, 26.1, 0.472, 1.092), neg: 165 (%0.01:%0.03), avgs: 1024 integrating with navgs=256 and tol=5.010e-01 104: dt: 2.000, sse: 654661.2 (0.357, 25.9, 0.470, 1.083), neg: 82 (%0.00:%0.01), avgs: 256 integrating with navgs=64 and tol=2.519e-01 105: dt: 3.105, sse: 653928.0 (0.357, 25.8, 0.469, 1.076), neg: 67 (%0.00:%0.01), avgs: 64 integrating with navgs=16 and tol=1.288e-01 106: dt: 2.292, sse: 653418.9 (0.357, 25.8, 0.469, 1.066), neg: 53 (%0.00:%0.01), avgs: 16 integrating with navgs=4 and tol=6.988e-02 107: dt: 1.047, sse: 653125.0 (0.357, 25.8, 0.469, 1.057), neg: 35 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-02 108: dt: 0.724, sse: 653027.9 (0.357, 25.8, 0.469, 1.047), neg: 36 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-02 109: dt: 0.064, sse: 653009.2 (0.357, 25.8, 0.469, 1.044), neg: 34 (%0.00:%0.00), avgs: 0 Removing remaining folds... tol=1.0e-01, sigma=0.5, host=athen, nav=64, nbrs=1, l_extern=10000.000, l_nlarea=100.000, l_corr=0.001, l_spring=0.005, l_dist=0.002 using quadratic fit line minimization nlarea/corr = 199999.984 integrating with navgs=64 and tol=2.519e-02 110: dt: 5.514, sse: 187681.1 (0.355, 25.8, 0.466, 1.062), neg: 85 (%0.00:%0.01), avgs: 64 111: dt: 1.735, sse: 184602.0 (0.355, 25.7, 0.466, 1.064), neg: 16 (%0.00:%0.00), avgs: 64 112: dt: 2.802, sse: 182518.2 (0.354, 25.7, 0.466, 1.068), neg: 16 (%0.00:%0.00), avgs: 64 113: dt: 3.973, sse: 180638.1 (0.355, 25.8, 0.466, 1.072), neg: 16 (%0.00:%0.00), avgs: 64 114: dt: 2.388, sse: 179649.5 (0.355, 25.8, 0.467, 1.075), neg: 20 (%0.00:%0.00), avgs: 64 115: dt: 3.088, sse: 178962.5 (0.355, 25.9, 0.467, 1.079), neg: 28 (%0.00:%0.00), avgs: 64 116: dt: 2.775, sse: 178633.9 (0.356, 26.0, 0.468, 1.082), neg: 36 (%0.00:%0.00), avgs: 64 117: dt: 0.831, sse: 178379.4 (0.356, 26.0, 0.468, 1.083), neg: 37 (%0.00:%0.00), avgs: 64 118: dt: 2.731, sse: 178093.2 (0.356, 26.1, 0.469, 1.087), neg: 52 (%0.00:%0.00), avgs: 64 119: dt: 0.606, sse: 177994.0 (0.356, 26.1, 0.469, 1.088), neg: 46 (%0.00:%0.00), avgs: 64 120: dt: 0.250, sse: 177987.2 (0.356, 26.1, 0.469, 1.088), neg: 48 (%0.00:%0.00), avgs: 64 integrating with navgs=16 and tol=1.288e-02 121: dt: 0.230, sse: 177851.4 (0.357, 26.1, 0.470, 1.089), neg: 46 (%0.00:%0.00), avgs: 16 122: dt: 0.031, sse: 177844.3 (0.357, 26.1, 0.470, 1.089), neg: 45 (%0.00:%0.00), avgs: 16 integrating with navgs=4 and tol=6.988e-03 123: dt: 0.013, sse: 177801.5 (0.357, 26.1, 0.470, 1.089), neg: 38 (%0.00:%0.00), avgs: 4 124: dt: 0.014, sse: 177784.2 (0.357, 26.1, 0.470, 1.089), neg: 32 (%0.00:%0.00), avgs: 4 125: dt: 0.014, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 4 integrating with navgs=1 and tol=4.419e-03 126: dt: 0.000, sse: 177779.8 (0.357, 26.1, 0.470, 1.089), neg: 31 (%0.00:%0.00), avgs: 1 integrating with navgs=0 and tol=3.125e-03 127: dt: 0.000, sse: 177746.4 (0.357, 26.1, 0.470, 1.089), neg: 30 (%0.00:%0.00), avgs: 0 128: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 129: dt: 0.000, sse: 177734.2 (0.357, 26.1, 0.470, 1.089), neg: 28 (%0.00:%0.00), avgs: 0 registration took 0.25 hours MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25_parea0.reg... registration took 0.25 hours [athens-b8-e8-56-47-e6-54:P00001613/FreeSurferClean/surf] ray%
-- 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 6, 2015, at 2:16 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
try doing:
setenv DIAG 0x04040
then run it again and send me the output
Bruce
On Fri, 6 Nov 2015, Razlighi, Qolamreza R. wrote:
Dear Bruce, Thank again for your kind help. I’m running version 5.1.0. Below is the output of the command. I also ran it with parea=0 but almost no change in the result (see the attachment-1) I also plot the sluci map for comparison (see the attachment-2) but still not satisfactory.[IMAGE][IMAGE] Best ~/Data/P00001613/FreeSurferClean/surf> mris_register -1 -curv -dist .25 lh.sphere ../../../P00001639/FreeSurferClean/surf/lh.sphere lh.sphere_dis25.reg treating target as a single subject's surface... using smoothwm curvature for final alignment l_dist = 0.250 $Id: mris_register.c,v 1.59 2011/03/02 00:04:33 nicks Exp $ $Id: mrisurf.c,v 1.693.2.7 2013/05/12 22:28:01 nicks Exp $ reading surface from lh.sphere... reading spherical surface ../../../P00001639/FreeSurferClean/surf/lh.sphere... curvature mean = -0.000, std = 1.000 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.inflated.H... curvature mean = 0.000, std = 0.566 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.sulc... curvature mean = -0.030, std = 0.282 computing parameterization for surface ../../../P00001639/FreeSurferClean/surf/lh.smoothwm... MRISregister() ------- max_passes = 4 min_degrees = 0.500000 max_degrees = 64.000000 nangles = 8 tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization complete_dist_mat 0 rms 0 smooth_averages 0 remove_neg 0 ico_order 0 which_surface 0 target_radius 0.000000 nfields 0 scale 0.000000 desired_rms_height -1.000000 momentum 0.950000 nbhd_size -10 max_nbrs 10 niterations 25 nsurfaces 0 SURFACES 3 flags 16 (10) use curv 16 no sulc 0 no rigid align 0 mris->nsize 1 mris->hemisphere 0 randomSeed 0 -------------------- tol=5.0e-01, sigma=0.0, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=1.000, l_dist=0.250 using quadratic fit line minimization 1 Reading lh.sulc curvature mean = 0.000, std = 0.583 curvature mean = 0.044, std = 0.847 curvature mean = 0.022, std = 0.852 Starting MRISrigidBodyAlignGlobal() d=32.00 min @ (8.00, 0.00, 0.00) sse = 90765.7, tmin=0.9987 d=16.00 min @ (-4.00, -4.00, 0.00) sse = 87371.6, tmin=1.5086 d=8.00 min @ (-2.00, 0.00, 0.00) sse = 86754.0, tmin=2.0493 d=4.00 min @ (1.00, 1.00, 0.00) sse = 86449.9, tmin=2.5976 d=2.00 min @ (0.00, -0.50, 0.50) sse = 86436.5, tmin=3.1563 d=1.00 min @ (0.00, 0.25, -0.25) sse = 86423.4, tmin=3.6949 MRISrigidBodyAlignGlobal() done 4.24 min curvature mean = 0.036, std = 0.923 curvature mean = 0.010, std = 0.931 curvature mean = 0.036, std = 0.951 curvature mean = 0.004, std = 0.967 curvature mean = 0.036, std = 0.962 curvature mean = 0.001, std = 0.984 2 Reading smoothwm curvature mean = -0.027, std = 0.269 tol=1.0e+00, sigma=0.5, host=unkno, nav=1024, nbrs=1, l_extern=10000.000, l_parea=0.100, l_nlarea=1.000, l_corr=0.050, l_spring=0.500, l_dist=0.250 using quadratic fit line minimization curvature mean = 0.090, std = 0.337 curvature mean = 0.059, std = 0.386 curvature mean = 0.091, std = 0.507 curvature mean = 0.020, std = 0.580 curvature mean = 0.091, std = 0.618 curvature mean = 0.011, std = 0.723 curvature mean = 0.093, std = 0.681 curvature mean = 0.004, std = 0.828 MRISregister() return, current seed 0 expanding nbhd size to 1 writing registered surface to lh.sphere_dis25.reg... -- 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 4, 2015, at 7:30 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
what version of FS are you using? Can you send me the output of the command? The distance term will prevent the curvatures from deforming too much. You can set it much smaller and see what happens if you want. There may also be an area constraint. Trying using -parea 0 also (or something small)
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
Hi Bruce, Thanks for reply. I thought after registering, the source surface's curvature map should be very similar to the target one. I don't see that here. The registered surface has pretty much the same curvature map only slightly shifted. Am I missing something here?
Best
On Nov 4, 2015, at 5:56 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> wrote:
Hi Ray
that sounds right. The way I visualize these is using nmovie (which I think we include in our distribution) and flipping back and forth between the different images showing the different surfaces/curv maps.
Which inaccuracy are you referring to?
cheers Bruce
On Wed, 4 Nov 2015, Razlighi, Qolamreza R. wrote:
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 [IMAGE] -- 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/listinf o/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.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurf er
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.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ 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.
_______________________________________________ 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. _______________________________________________ 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.
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Razlighi, Qolamreza R.