Quick update: this fix worked brilliantly. Many thanks for the help, J Richey
-----Original Message----- From: Douglas N Greve [mailto:[email protected]] Sent: Monday, July 18, 2011 5:20 PM To: John Richey Cc: 'Bruce Fischl'; '[email protected]' Subject: Re: [Freesurfer] Problem registering labels to native space In that case, try this instead doug # Create a register.dat to fsaverage/mni305 cd $SUBJECTS_DIR/subject/mri tkregister2 --s subject --targ orig.mgz \ --ixfm transforms/talairach.xfm --zero-cras \ --reg transforms/talairach.reg.dat \ --mov $FREESURFER_HOME/subjects/fsaverage/mri/orig.mgz \ --noedit # Create a register.dat to the 2mm version of fsaverage/mni305 mri_matrix_multiply \ -im $FREESURFER_HOME/subjects/fsaverage/mri.2mm/reg.2mm.dat \ -im transforms/talairach.reg.dat \ -om transforms/mni305.2mm.reg.dat # Check tkregister2 --reg transforms/mni305.2mm.reg.dat \ --mov $FREESURFER_HOME/subjects/fsaverage/mri.2mm/orig.mgz \ --surfs # Run label2label mri_label2label --srclabel mni305.2mm.label \ --trglabel $SUBJECTS_DIR/subject/labels/mni305.2mm.label \ --regmethod volume \ --reg $SUBJECTS_DIR/subject/mri/transforms/mni305.2mm.reg.dat \ --xfm-invert John Richey wrote: > Hi Doug, this is indeed an fsfast analysis. I used the fsaverage > subject included with the 5.1 distribution. Here is the code that I > used to produce the registration matrix for one subject. Note that I > did use mri_convert to change the .mgz files into .nii.gz > > The use of flirt (below) is more of an ad-hoc solution to register > each subject's anatomical image to fsaverage. I'm not wedded to this > method if there's a better alternative. > > bet ${SUBJECTS_DIR}/${SUBJECT}/mri/nu.nii.gz \ > ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet.nii.gz \ -f 0.3 -g 0 > > flirt -in ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet.nii.gz \ -ref > ${SUBJECTS_DIR}/fsaverage/mri/brain.nii.gz \ -out > ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet_reg.nii.gz \ -omat > ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet_reg.mat \ -bins 256 -cost > corratio -searchrx -90 90 -searchry -90 90 \ -searchrz -90 90 -dof 12 > -interp trilinear > > cd $SUBJECTS_DIR > tkregister2 --targ $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.nii.gz \ > --mov $SUBJECTS_DIR/fsaverage/mri/brain.nii.gz \ --fsl > $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.mat \ --reg > $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.dat \ --noedit > > > > -----Original Message----- > From: Douglas N Greve [mailto:[email protected]] > Sent: Monday, July 18, 2011 3:11 PM > To: John Richey > Cc: 'Bruce Fischl'; '[email protected]' > Subject: Re: [Freesurfer] Problem registering labels to native space > > How did you create it? What is the actual "fsaverage space"? Is this > an FSFAST analysis? > doug > > John Richey wrote: > >> Hi Doug, the .dat file that I mention below had already been >> converted >> > into > >> a fs matrix using tkregister2. >> >> -----Original Message----- >> From: Douglas N Greve [mailto:[email protected]] >> Sent: Monday, July 18, 2011 12:38 PM >> To: John Richey >> Cc: 'Bruce Fischl'; '[email protected]' >> Subject: Re: [Freesurfer] Problem registering labels to native space >> >> You can't use a flirt matrix with mri_label2label. You can convert a >> flirt matrix to the proper format with tkregister2 using the --fsl >> option (this will allow you to check the reg as well). >> >> doug >> >> John Richey wrote: >> >> >>> Hi Bruce, >>> >>> The labels that I have fsaverage space are 3D/volume-based and not >>> surface-based labels. >>> >>> >>> -----Original Message----- >>> From: Bruce Fischl [mailto:[email protected]] >>> Sent: Monday, July 18, 2011 12:00 PM >>> To: John Richey >>> Cc: '[email protected]' >>> Subject: RE: [Freesurfer] Problem registering labels to native space >>> >>> Hi John >>> >>> why are you using an affine volume registration for fsaverage? Why >>> not use surface registration? >>> >>> Bruce >>> >>> >>> On Mon, 18 Jul 2011, John Richey wrote: >>> >>> >>> >>> >>>> Hi Bruce - thanks for the reply. >>>> >>>> Here is my command line: >>>> >>>> mri_label2label --srcsubject fsaverage \ --trgsubject ${SUBJECT} >>>> --srclabel $SUBJECTS_DIR/group_analysis/${CONTRAST}/${LABEL} \ >>>> --trglabel $SUBJECTS_DIR/${SUBJECT}/label/${CONTRAST}_${LABEL} \ >>>> --regmethod volume \ --reg $SUBJECTS_DIR/$SUBJECT/mri/brain_reg.dat >>>> >>>> Where "brain_reg.dat" is the subject-specific matrix I created >>>> through FLIRT. However, as I mentioned, disincluding this matrix >>>> does not seem to change the result much. >>>> >>>> One other problem I encounter... Occasionally when I view the >>>> label in native space, the label itself seems to be eroded on a >>>> voxel-wise >>>> >>>> >>>> >>> basis. >>> >>> >>> >>>> That is, the label essentially appears as a series of >>>> non-contiguous dots (which if dialated would become contiguous). >>>> >>>> >>>> >>>> >>>> >>>> >>>> -----Original Message----- >>>> From: Bruce Fischl [mailto:[email protected]] >>>> Sent: Monday, July 18, 2011 11:20 AM >>>> To: John Richey >>>> Cc: '[email protected]' >>>> Subject: Re: [Freesurfer] Problem registering labels to native >>>> space >>>> >>>> Hi John >>>> >>>> can you email us your label2label cmdline? >>>> >>>> cheers >>>> Bruce >>>> >>>> >>>> On Mon, 18 Jul 2011, John Richey >>>> wrote: >>>> >>>> >>>> >>>> >>>>> Dear freesurfer experts, >>>>> >>>>> I have a series of labels corresponding to clusters of significant >>>>> activation in fsaverage space. >>>>> >>>>> I want to map these back to native space using mri_label2label, >>>>> however when I view the individual (native) labels, the >>>>> registration is way off and the labels do not map into plausible space. >>>>> >>>>> I have tried using talairach space as an intemediary (by omitting >>>>> the --reg flag in mri_label2label), as well as using fsl tools >>>>> (FLIRT) to create a registration matrix but unfortunately I cannot >>>>> seem to get the label into the right space to query my ces maps in >>>>> >>>>> >> funcroi-sess. >> >> >>>>> Have others encountered this problem? And how might I create a >>>>> more robust registration between fsaverage and native space to >>>>> facilitate the spatial transformation of my labels? >>>>> >>>>> Thanks very much in advance, >>>>> John >>>>> _______________________________________________ >>>>> Freesurfer mailing list >>>>> [email protected] >>>>> 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 >>> [email protected] >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >>> >>> >>> >>> >> -- >> Douglas N. Greve, Ph.D. >> MGH-NMR Center >> [email protected] >> 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 >> >> >> >> > > -- > Douglas N. Greve, Ph.D. > MGH-NMR Center > [email protected] > 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 > > > -- Douglas N. Greve, Ph.D. MGH-NMR Center [email protected] 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 _______________________________________________ Freesurfer mailing list [email protected] https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
