Hi Eric,
Yes, some people are, there was even a short tutorial/howto posted on this very list (maybe you can find it in the archives).Message: 1 Date: Wed, 10 May 2006 09:37:43 -0400 From: "Faden, Eric (NIH/NIMH) [F]" <[EMAIL PROTECTED]> Subject: [Freesurfer] Skull Stripping a Macaque To: <freesurfer@nmr.mgh.harvard.edu> Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="iso-8859-1"Hey all, Is anyone using freesurfers mri_watershed to strip macaque skulls?
Inlined after your mail you can find the script I use for skull stripping. The take home message from the howto, at least for me, was to try different sets of parameters with "-surf_debug" to nicely visualize the results. Unfortunately I got the best results so far only with an additional step of manual editing of the brainmask, as my scans suffer from inhomogenities (surface coil used for transmit and receive). For the manual edit steps I recommend the fine skull stripping howto found somewhere at the freesurfer wiki. One thing that caused slight hickups for me, was r´the fact that the erase funtion of the volume brush does not write zeros but ones, which messed up the segmentation later on, as the voxels with a value of one got much brighter by the normalisation step, but I digress...If so what options are you using?
I am currently trying to use it but am getting really bad results.
Care to compare brainmasks to see what you consider bad?
I am also open to suggestions of other programs to try. -Eric
Ahoi
Sebastian
------Cut and paste from here-----
#! /bin/csh -f
# use
tkmedit to mark critical whitematter points
#setenv
FIX_VERTEX_AREA
# set the source directory...
set src_dir = ../mri
# set the target directory and file_name
set
targ_dir = ../mri
#set reg =
rotate_conf_2_atlas.dat
#set targ_vol =
rawavg.mgz
#set controlpoints =
../tmp/control.dat
set in_vol = ${src_dir}/nu.mgz
#orig.mgz
set t1 = ${targ_dir}/T1.mgz
# do
the skull stripping
# repeat with
-surf_debug until the parameters are okay, the run without...
#mri_watershed
-surf_debug -atlas ${targ_dir}/${t1} ${targ_dir}/brainmask.auto.mgz
#mri_watershed -surf_debug
-atlas -wat+temp ${targ_dir}/${t1} ${targ_dir}/brainmask.auto.mgz
#mri_watershed -surf_debug -h 25
-wat+temp ${targ_dir}/${t1} ${targ_dir}/brainmask.auto.mgz
#mri_watershed -h 10 -wat -n
${targ_dir}/${t1} ${targ_dir}/brainmask.auto.mgz
mri_watershed -atlas ${targ_dir}/${t1}
${targ_dir}/brainmask.auto.mgz
# create the brainmask
##cp ${targ_dir}/brainmask.auto.mgz ${targ_dir}/brainmask.mgz
# check
results...
tkmedit -f
${targ_dir}/brainmask.mgz -aux ${targ_dir}/T1.mgz
#cp ${t1} ${targ_dir}/brain.mgz
# does a register file already exist?
#if !(-e ${targ_dir}/${reg}) then
# cp ./${reg} ${targ_dir}/${reg}
#endif
# convert to cor? or turn this sucker so that
it works with mgz?
#TODO: normalize with control points...
----Cut and paste until here--------------- -- Sebastian Moeller Tel.: 04 21 - 2 18 - 78 38 oder 96 91 Fax.: 04 21 - 2 18 - 90 04 GSM: 01 62 - 3 25 45 59 [EMAIL PROTECTED] AG Kreiter / FB 2 Institut fuer Hirnforschung III Abteilung Theoretische Neurobiologie Universitaet Bremen Biogarten Hochschulring 16a Postfach 33 04 40 28359 Bremen _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
