28 Sep
2012
28 Sep
'12
10:18 a.m.
hi Gari, looking at this I don't really see a problem. Yes, the t-test is very significant, but I would expect that from a paired t looking across different methods. the size of the effect is very small, about 1.5%. Do you see this as a problem? doug
On 09/28/2012 05:14 AM, Garikoitz Lerma-Usabiaga wrote:
Hi Doug, I think that the differences are significant enough, but just in case please find attached an excel file with the data and the t-test.
Some of the ROI-s are small (Triangularis, Operculum, Orbitalis and then all together in the IFG). I attached images for each of them for reference.
If in any case you assume that there are going to be differences, which method would you recommend and why?
Thanks! Gari
PS after mri_preproc I didn't smooth the surface as in the qdec example, ¿should I do that?
On Fri, Sep 28, 2012 at 4:24 AM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
OK, got it (finally). I would not expect the answers to be identical, but I would expect them to be "close". How close are they? If the ROI is very small, there might be appreciable differences between the two. doug On 9/27/12 6:26 PM, Garikoitz Lerma-Usabiaga wrote:Hi, You are right, this was one of the methods. The second one is as follows (I copy the repeated steps):For a given ROI, you have a nifti file (roi.nii) in 2mm space defined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mghI run mris_preproc for every subject to Colin27: mris_preproc --out /subject/surf/ROI_toColin27.mgh --target Colin27 --hemi lh --meas thickness --s subject mri_segstats --seg /Colin27/surf/lh.roi.mgh --in /subject/surf/ROI_to_Colin127.mgh --sum subject/stats/target_lh.ROI.stats Thanks for your help, Gari Note: You said: You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label Just in case, this is how I run it: * mri_cor2label --i /subjects/Colin27/lh.ROI.mgh --surf Colin27 lh --id 1 --l /subjects/Colin27/lh.ROI.label * mri_label2label --srclabel /subjects/Colin27/lh.ROI.label --srcsubject Colin27 --trgsubject subject --trglabel lh.ROI_Colin27_to_bert --regmethod surface --hemi lh On 26/09/2012, at 23:56, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> wrote:Sorry, I'm still a little lost. Let me see if I have it right: For a given ROI, you have a nifti file (roi.nii) in 2mm space defined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mgh You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label You run mri_label2label on lh.roi.label to map it into an individual subject You run mris_anatomical_stats on the individual lh.roi.label file Does this describe one of your processing streams correctly? If so, what is the other processing stream (given in this kind of detail)? doug On 09/25/2012 04:43 PM, Garikoitz Lerma-Usabiaga wrote:I have both. In origin, the ROI-s are 2mm volume based .nii that I convert to FS .mgh with register.dat I obtained with bbregister. -- In the case of label2label method I create the surface version with mri_vol2surf (and the correspondence is perfect), and then use mri_label2label in order to get the CT averages. -- In the case of surf2surf, I use the volume based .mgh with mri_preproc (mri_surf2surf) thanks! Gari On Tue, Sep 25, 2012 at 5:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu>> wrote: So you have ROIs in Colin27 space and you want to map them into your CT space and compute average CT intensities over them? Are the ROIs volume-based or surface-based? doug On 09/25/2012 05:09 AM, Garikoitz Lerma-Usabiaga wrote: Hi Doug, many thanks for your answer, I try to detail it more: 1.- bbregister is registering two Colin27 brains. First one is a 2mm found in spm/canonical/single_subject_T1.nii, and second one is the FS_spm_Canonical found in SurfRend distribution. It was in .COR format and I run recon-all to have a modern version of it. As they are almost the same brain the registration matrix is quite simple and the registration was perfect. This way I had exactly the same ROI-s for Colin27 in SPM and in FS (it did not work that well for fsaverage, that's why I took this approach). 2.- In any case, I have several ROI-s in Colin27 space in FS, and then when I try to obtain average CT values from my subjects I use these 2 different approaches: label2label and surf2surf. mri_anatomical_stats is run on the labels created by mri_label2label. If you think it will be of help I can send the detailed command line calls, thanks! Gari On Sat, Sep 22, 2012 at 12:47 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: Hi Gari, I'll need more information. I can't tell what you are doing. Eg, what is being registered with BBR? what is mris_anatomical_stats run on? Where does Colin27 come into it and why? doug On 09/19/2012 06:15 PM, Garikoitz Lerma-Usabiaga (Gari) wrote:Dear Freesurfers, We've been comparing 2 different methods to obtain CT averagesfor 4 Volume ROI-s (volume ROI-s were in Colin27 space).We've used the 40 buckner subjects for the comparisons. METHOD L2L: bbregister>> mri_vol2surf (mgh)>>mri_binarize (mgh)>> mri_cor2label (label)>> mri_label2label>> mris_anatomical_statsMETHOD S2S: bbregister>> mri_vol2surf>> mri_preproc(mri_surf2surf)>> mri_segstatsAfter performing t-test-s over the results, we can observe thatthe results are in many cases different.-- Is there a way to choose the "best" method? Which one shouldwe use for our work and why?Many thanks again for your help, Gari PD We've tabulated a third method as well: METHOD SurfRend: Surfrend (.w)>> mri_surf2surf(.mgh)>>mri_binarize(.mgh)>> mri_cor2label (.label)>> mris_anatomical_stats_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu><mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>>https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> Phone Number: 617-724-2358 <tel:617-724-2358> <tel:617-724-2358 <tel:617-724-2358>> Fax: 617-726-7422 <tel:617-726-7422> <tel:617-726-7422 <tel:617-726-7422>> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto: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. -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu> Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
29 Sep
29 Sep
5:17 p.m.
New subject: CT Method Comparison
Hi Doug, you are right, we are very happy with the results, we were just trying to be as conservative as possible and testing different methods and regions before making any decision. We are going to introduce one of the methods in a stream for several studies and we wanted to know if you have any recommendation. As the results are not exactly the same maybe you think one of those is better than the other for any reason.
For example, considering that the accuracy between methods is similar, could we use the data in Qdec (with Colin27) with the surf2surf method? I understand that we couldn't do that with the label2label method.
Many thanks again for your help! Gari
On Fri, Sep 28, 2012 at 4:18 PM, Douglas N Greve greve@nmr.mgh.harvard.eduwrote:
hi Gari, looking at this I don't really see a problem. Yes, the t-test is very significant, but I would expect that from a paired t looking across different methods. the size of the effect is very small, about 1.5%. Do you see this as a problem? doug
On 09/28/2012 05:14 AM, Garikoitz Lerma-Usabiaga wrote:
Hi Doug,
I think that the differences are significant enough, but just in case please find attached an excel file with the data and the t-test.
Some of the ROI-s are small (Triangularis, Operculum, Orbitalis and then all together in the IFG). I attached images for each of them for reference.
If in any case you assume that there are going to be differences, which method would you recommend and why?
Thanks! Gari
PS after mri_preproc I didn't smooth the surface as in the qdec example, żshould I do that?
On Fri, Sep 28, 2012 at 4:24 AM, Douglas Greve <greve@nmr.mgh.harvard.edu<mailto: greve@nmr.mgh.harvard.**edu greve@nmr.mgh.harvard.edu>> wrote:
OK, got it (finally). I would not expect the answers to be identical, but I would expect them to be "close". How close are they? If the ROI is very small, there might be appreciable differences between the two. doug On 9/27/12 6:26 PM, Garikoitz Lerma-Usabiaga wrote:Hi, You are right, this was one of the methods. The second one is as follows (I copy the repeated steps): For a given ROI, you have a nifti file (roi.nii) in 2mm spacedefined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mghI run mris_preproc for every subject to Colin27: mris_preproc --out /subject/surf/ROI_toColin27.**mgh --target Colin27 --hemi lh --meas thickness --s subject mri_segstats --seg /Colin27/surf/lh.roi.mgh --in /subject/surf/ROI_to_Colin127.**mgh --sum subject/stats/target_lh.ROI.**stats Thanks for your help, Gari Note: You said: You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label Just in case, this is how I run it: * mri_cor2label --i /subjects/Colin27/lh.ROI.mgh --surf Colin27 lh --id 1 --l /subjects/Colin27/lh.ROI.label * mri_label2label --srclabel /subjects/Colin27/lh.ROI.label --srcsubject Colin27 --trgsubject subject --trglabel lh.ROI_Colin27_to_bert --regmethod surface --hemi lh On 26/09/2012, at 23:56, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu>>>wrote:
Sorry, I'm still a little lost. Let me see if I have it right:For a given ROI, you have a nifti file (roi.nii) in 2mm space defined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mgh You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label You run mri_label2label on lh.roi.label to map it into an individual subject You run mris_anatomical_stats on the individual lh.roi.label file Does this describe one of your processing streams correctly? If so, what is the other processing stream (given in this kind of detail)? doug On 09/25/2012 04:43 PM, Garikoitz Lerma-Usabiaga wrote:I have both. In origin, the ROI-s are 2mm volume based .nii that I convert to FS .mgh with register.dat I obtained with bbregister. -- In the case of label2label method I create the surface version with mri_vol2surf (and the correspondence is perfect), and then use mri_label2label in order to get the CT averages. -- In the case of surf2surf, I use the volume based .mgh with mri_preproc (mri_surf2surf) thanks! Gari On Tue, Sep 25, 2012 at 5:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu><mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>>>wrote:
So you have ROIs in Colin27 space and you want to map them into your CT space and compute average CT intensities over them? Are the ROIs volume-based or surface-based? doug On 09/25/2012 05:09 AM, Garikoitz Lerma-Usabiaga wrote: Hi Doug, many thanks for your answer, I try to detail it more: 1.- bbregister is registering two Colin27 brains. First one is a 2mm found in spm/canonical/single_subject_**T1.nii, and second one is the FS_spm_Canonical found in SurfRend distribution. It was in .COR format and I run recon-all to have a modern version of it. As they are almost the same brain the registration matrix is quite simple and the registration was perfect. This way I had exactly the same ROI-s for Colin27 in SPM and in FS (it did not work that well for fsaverage, that's why I took this approach). 2.- In any case, I have several ROI-s in Colin27 space in FS, and then when I try to obtain average CT values from my subjects I use these 2 different approaches: label2label and surf2surf. mri_anatomical_stats is run on the labels created by mri_label2label. If you think it will be of help I can send the detailed command line calls, thanks! Gari On Sat, Sep 22, 2012 at 12:47 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu><mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>>>>wrote:
Hi Gari, I'll need more information. I can't tell what you are doing. Eg, what is being registered with BBR? what is mris_anatomical_stats run on? Where does Colin27 come into it and why? doug On 09/19/2012 06:15 PM, Garikoitz Lerma-Usabiaga (Gari) wrote:Dear Freesurfers, We've been comparing 2 different methods to obtain CT averagesfor 4 Volume ROI-s (volume ROI-s were in Colin27 space).We've used the 40 buckner subjects for the comparisons. METHOD L2L: bbregister>> mri_vol2surf (mgh)>>mri_binarize (mgh)>> mri_cor2label (label)>> mri_label2label>> mris_anatomical_statsMETHOD S2S: bbregister>> mri_vol2surf>> mri_preproc(mri_surf2surf)>> mri_segstatsAfter performing t-test-s over the results, we can observe thatthe results are in many cases different.-- Is there a way to choose the "best" method? Which one shouldwe use for our work and why?Many thanks again for your help, Gari PD We've tabulated a third method as well: METHOD SurfRend: Surfrend (.w)>> mri_surf2surf(.mgh)>>mri_binarize(.mgh)>> mri_cor2label (.label)>> mris_anatomical_stats______________________________**_________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu>>
<mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu><mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu>>
https://mail.nmr.mgh.harvard.**edu/mailman/listinfo/**freesurfer<https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu><mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>>> Phone Number: 617-724-2358 <tel:617-724-2358> <tel:617-724-2358 <tel:617-724-2358>> Fax: 617-726-7422 <tel:617-726-7422> <tel:617-726-7422 <tel:617-726-7422>> Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting><http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting><http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>FileDrop: www.nmr.mgh.harvard.edu/**facility/filedrop/index.html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html><http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html><http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>______________________________**_________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu><mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu><mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.**harvard.edu<Freesurfer@nmr.mgh.harvard.edu>>
https://mail.nmr.mgh.harvard.**edu/mailman/listinfo/**freesurfer<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<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. -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu><mailto:greve@nmr.mgh.harvard.**edu <greve@nmr.mgh.harvard.edu>> Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting><http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>FileDrop: www.nmr.mgh.harvard.edu/**facility/filedrop/index.html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html><http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.**edu<greve@nmr.mgh.harvard.edu>Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.**harvard.edu/fswiki/**BugReporting<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>FileDrop: www.nmr.mgh.harvard.edu/**facility/filedrop/index.html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.**edu/facility/filedrop/index.**html<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/**fswiki/BugReportinghttp://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/**facility/filedrop/index.htmlhttp://www.nmr.mgh.harvard.edu/facility/filedrop/index.html
1 Oct
1 Oct
6:08 p.m.
New subject: CT Method Comparison
Hi Gari, you can use the data from the surf2surf in qdec/mri_glmfit. If you are using qdec, you might want to let qdec generate the file for you to make sure they are consistent. It is just running mris_preproc and should give an identical file, but just to make sure.
doug
On 09/29/2012 05:17 PM, Garikoitz Lerma-Usabiaga wrote:
Hi Doug, you are right, we are very happy with the results, we were just trying to be as conservative as possible and testing different methods and regions before making any decision. We are going to introduce one of the methods in a stream for several studies and we wanted to know if you have any recommendation. As the results are not exactly the same maybe you think one of those is better than the other for any reason.
For example, considering that the accuracy between methods is similar, could we use the data in Qdec (with Colin27) with the surf2surf method? I understand that we couldn't do that with the label2label method.
Many thanks again for your help! Gari
On Fri, Sep 28, 2012 at 4:18 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:
hi Gari, looking at this I don't really see a problem. Yes, the t-test is very significant, but I would expect that from a paired t looking across different methods. the size of the effect is very small, about 1.5%. Do you see this as a problem? doug On 09/28/2012 05:14 AM, Garikoitz Lerma-Usabiaga wrote: Hi Doug, I think that the differences are significant enough, but just in case please find attached an excel file with the data and the t-test. Some of the ROI-s are small (Triangularis, Operculum, Orbitalis and then all together in the IFG). I attached images for each of them for reference. If in any case you assume that there are going to be differences, which method would you recommend and why? Thanks! Gari PS after mri_preproc I didn't smooth the surface as in the qdec example, żshould I do that? On Fri, Sep 28, 2012 at 4:24 AM, Douglas Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: OK, got it (finally). I would not expect the answers to be identical, but I would expect them to be "close". How close are they? If the ROI is very small, there might be appreciable differences between the two. doug On 9/27/12 6:26 PM, Garikoitz Lerma-Usabiaga wrote: Hi, You are right, this was one of the methods. The second one is as follows (I copy the repeated steps): For a given ROI, you have a nifti file (roi.nii) in 2mm space defined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mgh I run mris_preproc for every subject to Colin27: mris_preproc --out /subject/surf/ROI_toColin27.mgh --target Colin27 --hemi lh --meas thickness --s subject mri_segstats --seg /Colin27/surf/lh.roi.mgh --in /subject/surf/ROI_to_Colin127.mgh --sum subject/stats/target_lh.ROI.stats Thanks for your help, Gari Note: You said: You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label Just in case, this is how I run it: * mri_cor2label --i /subjects/Colin27/lh.ROI.mgh --surf Colin27 lh --id 1 --l /subjects/Colin27/lh.ROI.label * mri_label2label --srclabel /subjects/Colin27/lh.ROI.label --srcsubject Colin27 --trgsubject subject --trglabel lh.ROI_Colin27_to_bert --regmethod surface --hemi lh On 26/09/2012, at 23:56, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: Sorry, I'm still a little lost. Let me see if I have it right: For a given ROI, you have a nifti file (roi.nii) in 2mm space defined on colin27. This ROI volume has a voxel value == 1 if it is in the ROI or 0 if it is out. You have a registration between roi.nii and orig.mgz You run mri_vol2surf on roi.nii to get lh.roi.mgh You run mri_cor2label (with --surf option) to convert lh.roi.mgh into lh.roi.label You run mri_label2label on lh.roi.label to map it into an individual subject You run mris_anatomical_stats on the individual lh.roi.label file Does this describe one of your processing streams correctly? If so, what is the other processing stream (given in this kind of detail)? doug On 09/25/2012 04:43 PM, Garikoitz Lerma-Usabiaga wrote: I have both. In origin, the ROI-s are 2mm volume based .nii that I convert to FS .mgh with register.dat I obtained with bbregister. -- In the case of label2label method I create the surface version with mri_vol2surf (and the correspondence is perfect), and then use mri_label2label in order to get the CT averages. -- In the case of surf2surf, I use the volume based .mgh with mri_preproc (mri_surf2surf) thanks! Gari On Tue, Sep 25, 2012 at 5:57 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> wrote: So you have ROIs in Colin27 space and you want to map them into your CT space and compute average CT intensities over them? Are the ROIs volume-based or surface-based? doug On 09/25/2012 05:09 AM, Garikoitz Lerma-Usabiaga wrote: Hi Doug, many thanks for your answer, I try to detail it more: 1.- bbregister is registering two Colin27 brains. First one is a 2mm found in spm/canonical/single_subject_T1.nii, and second one is the FS_spm_Canonical found in SurfRend distribution. It was in .COR format and I run recon-all to have a modern version of it. As they are almost the same brain the registration matrix is quite simple and the registration was perfect. This way I had exactly the same ROI-s for Colin27 in SPM and in FS (it did not work that well for fsaverage, that's why I took this approach). 2.- In any case, I have several ROI-s in Colin27 space in FS, and then when I try to obtain average CT values from my subjects I use these 2 different approaches: label2label and surf2surf. mri_anatomical_stats is run on the labels created by mri_label2label. If you think it will be of help I can send the detailed command line calls, thanks! Gari On Sat, Sep 22, 2012 at 12:47 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>>> wrote: Hi Gari, I'll need more information. I can't tell what you are doing. Eg, what is being registered with BBR? what is mris_anatomical_stats run on? Where does Colin27 come into it and why? doug On 09/19/2012 06:15 PM, Garikoitz Lerma-Usabiaga (Gari) wrote: Dear Freesurfers, We've been comparing 2 different methods to obtain CT averages for 4 Volume ROI-s (volume ROI-s were in Colin27 space). We've used the 40 buckner subjects for the comparisons. METHOD L2L: bbregister>> mri_vol2surf (mgh)>> mri_binarize (mgh)>> mri_cor2label (label)>> mri_label2label>> mris_anatomical_stats METHOD S2S: bbregister>> mri_vol2surf>> mri_preproc (mri_surf2surf)>> mri_segstats After performing t-test-s over the results, we can observe that the results are in many cases different. -- Is there a way to choose the "best" method? Which one should we use for our work and why? Many thanks again for your help, Gari PD We've tabulated a third method as well: METHOD SurfRend: Surfrend (.w)>> mri_surf2surf(.mgh)>> mri_binarize(.mgh)>> mri_cor2label (.label)>> mris_anatomical_stats _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>>> Phone Number: 617-724-2358 <tel:617-724-2358> <tel:617-724-2358 <tel:617-724-2358>> Fax: 617-726-7422 <tel:617-726-7422> <tel:617-726-7422 <tel:617-726-7422>> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> <mailto:Freesurfer@nmr.mgh.harvard.edu <mailto: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. -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422 <tel:617-726-7422> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> <mailto:greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>> Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html> -- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu> Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting <http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html <http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
5080
Age (days ago)
5083
Last active (days ago)
freesurfer@nmr.mgh.harvard.edu
2 comments
2 participants
participants (2)
-
Douglas N Greve -
Garikoitz Lerma-Usabiaga