no, you don't use brainmask as input, you just put it in the subject's mri dir. recon-all should then recognize that it is different than brainmask.auto.mgz, assume it has been manually edited, and preserve those edits On Tue, 18 Jul 2017, Michelle VanTieghem wrote:
Hi, Yes, I am trying to use skull stripping from another package, since -gcut option did not work on our data for appropriate skull stripping, and as a result there is too much pial matter included as white matter.
a) Is brainmask.mgz an appropriate input file for recon-all -all? for example, should I run "recon-all -all -i brainmask.mgz -s SB001" ? or, am I still using the original .dcm file as input, and merely putting a copy of my skull stripped brain into the SUB/mri directory that was created in the previous processing attempt?
b) I am not understanding why i will still run skull stripping in the script here, if it is already done. Is the brainmask.auto.mgz generated just another duplicate copy that will be ignored for the actual segmentations? Will this still perform intensity normalization and other auto-recon1 steps normally?
Thank you, Michelle
On Tue, Jul 18, 2017 at 12:44 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: Hi Michelle
can you cc the list so others can answers? I'm also no longer sure which thread this is - were you trying to use the skull stripping from another package? If so, convert that file to be named brainmask.mgz, then do *not* include the -noskullstrip in your command line. Let us do the skull strippping, which will create a brainmask.auto.mgz, but should preserve the brainmask.mgz (since it is different than the .auto file) cheers Bruce On Tue, 18 Jul 2017, Michelle VanTieghem wrote: Hi Bruce, Thanks for the information. Sorry I had changed my mail settings so I wasn’t seeing your responses until now. I had tried to run all, but just skip skull strip step, as well as the talaraich check step since we saw errors with that too. The command I used is follows: recon-all -all -3T -mprage -notal-check -noskullstrip-i/vega/psych/users/mrv2115/fsdata/ORIG_AFNI_SKULLSTRIP/SB001/mprage_deobli qu e_brain.nii -s SB001 However, I am getting the following error during the registration step (see below). It looks like the mask volume brainmask.mgz hasn't been generated, but i don't know why. You suggested to convert my afni skull stripped brain to brain mask.mgz and place in the MRI folder. but am I still going to run the steps of autorecon-1 ? (intensity normalization, etc)? Does recon-all take inputs of already existing mri/brainmask.mgz files? It seems like it just needs one file and then re-generates a new directory with mri, scripts, ect. I would appreciate more detail on how to implement this! Thank you, Michelle ERROR: ======= NUMBER OF OPENMP THREADS = 1 ======= aligning to atlas containing skull, setting unknown_nbr_spacing = 3 using MR volume brainmask.mgz to mask input volume... reading 1 input volumes... logging results to talairach.log reading '/vega/psych/app/freesurfer/average/RB_all_2008-03-26.gca'... average std = 6.9 using min determinant for regularization = 4.7 0 singular and 1812 ill-conditioned covariance matrices regularized reading 'nu.mgz'...mghRead(/vega/psych/users/mrv2115/fsdata/processed/SB001/mri/brainmask.mgz,
-1): could not $ mri_em_register: could not open mask volume brainmask.mgz. Linux bogota.cc.columbia.edu 2.6.32-696.el6.x86_64 #1 SMP Tue Feb 21 00:53:17 EST 2017 x86_$ recon-all -s SB001 exited with ERRORS at Tue Jul 18 11:50:24 EDT 2017 ——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.
-- Michelle VanTieghem PhD student in Psychology Developmental Affective Neuroscience Lab Columbia University mrv2115@columbia.edu