Hello, I have a question about how to get freesurfer to work correctly with images that have problematic contrast.

http://brainimaging.waisman.wisc.edu/~perlman/images/1280_2_MPRAGE_sag.png


Many of the images that look like this are processed just fine, but some of them fail.  When they fail, it looks like this in the log (see pasted log text below).
I looked at the link that it told me to check but it was an invalid page in the wiki:
http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Talairach
I tracked down this page instead:
http://surfer.nmr.mgh.harvard.edu/fswiki/FreeSurferTalairach

On that page it suggests:
When you see a big bright neck region, an option is to edit the volume to erase the bright neck region and run mritotal on the edited volume. It usually works by a simple editing, setting all the region to zero where y > certain value (around 170). This worked better than using "brain" volume.

I thresholded one of my trouble images with this command:
poblano:/study/gates/data/mri/preprocessed/VIPR_IR/MPRAGE$ 3dcalc -a 1221_2_MPRAGE_reorient.nii.gz  -expr 'step(a-1000)*1000 + step(1000-a)*a' -prefix 1221_2_MPRAGE_reorient_clip1000.nii.gz
++ 3dcalc: AFNI version=AFNI_2011_12_21_1014 (Mar 31 2014) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./1221_2_MPRAGE_reorient_clip1000.nii.gz

The image certainly looked better after that.

http://brainimaging.waisman.wisc.edu/~perlman/images/1280_2_MPRAGE_sag_trim1000.png

I tried it again with the command:
poblano:/study/gates/data/mri/preprocessed/VIPR_IR/MPRAGE/freesurfer_nu_test$ recon-all -i /study/gates/data/mri/preprocessed/VIPR_IR/MPRAGE/1221_2_MPRAGE_reorient_clip1000.nii.gz -subject 1221_2_clip1000 -all

but the result was the same.
The talairach.xfm matrix is this:

Transform_Type = Linear;
Linear_Transform =
 0.755387127399445 -0.110331535339355 -0.538617014884949 -24.9059925079346
 0.544680178165436 0.0255524087697268 0.75865626335144 15.3778610229492
 -0.0748601630330086 -0.927396953105927 0.0849819406867027 -111.666954040527;

which definitely doesn't look right.

We tried turning up the --nuiterations option to a much higher value, but this step doesn't run until after the Talairach transform so it is unable to help that process.  (Wouldn't it make more sense to swap that order?)

I was wondering if you could give me any advice on other steps I could try, to get this brain to process with recon_all?

Thanks!







INFO: Attempting MINC mritotal to perform Talairach align
#--------------------------------------------
#@# Talairach Mon Apr 14 10:54:57 CDT 2014
/study1/gates/data/mri/analysis/SA/freesurfer/test_mprage_fail/1221_2/mri

 mri_nu_correct.mni --n 1 --proto-iters 1000 --distance 50 --no-rescale --i orig.mgz --o orig_nu.mgz 


 talairach --i orig_nu.mgz --xfm transforms/talairach.auto.xfm 

/study1/gates/data/mri/analysis/SA/freesurfer/test_mprage_fail/1221_2/mri
/apps/x86_64_sci6/freesurfer-latest/bin/talairach
--i orig_nu.mgz --xfm transforms/talairach.auto.xfm
$Id: talairach,v 1.7 2011/03/02 20:16:40 nicks Exp $
Linux c114.keck.waisman.wisc.edu 2.6.32-431.5.1.el6.x86_64 #1 SMP Tue Feb 11 13:30:01 CST 2014 x86_64 x86_64 x86_64 GNU/Linux
Mon Apr 14 10:55:50 CDT 2014
tmpdir is transforms/tmp.talairach.30831
/study1/gates/data/mri/analysis/SA/freesurfer/test_mprage_fail/1221_2/mri
mri_convert orig_nu.mgz transforms/tmp.talairach.30831/src.mnc
mri_convert orig_nu.mgz transforms/tmp.talairach.30831/src.mnc 
$Id: mri_convert.c,v 1.179.2.7 2012/09/05 21:55:16 mreuter Exp $
reading from orig_nu.mgz...
TR=4.55, TE=0.00, TI=0.00, flip angle=0.00
i_ras = (-1, 0, 0)
j_ras = (0, 0, -1)
k_ras = (0, 1, 0)
writing to transforms/tmp.talairach.30831/src.mnc...
--------------------------------------------
mritotal -verbose -debug -clobber -modeldir /apps/x86_64_sci6/freesurfer-latest/mni/bin/../share/mni_autoreg -protocol icbm transforms/tmp.talairach.30831/src.mnc transforms/talairach.auto.xfm
 
 
Mon Apr 14 10:56:11 CDT 2014
talairach done

 cp transforms/talairach.auto.xfm transforms/talairach.xfm 

#--------------------------------------------
#@# Talairach Failure Detection Mon Apr 14 10:56:12 CDT 2014
/study1/gates/data/mri/analysis/SA/freesurfer/test_mprage_fail/1221_2/mri

 talairach_afd -T 0.005 -xfm transforms/talairach.xfm 

ERROR: talairach_afd: Talairach Transform: transforms/talairach.xfm ***FAILED*** (p=0.0000, pval=0.0000 < threshold=0.0050)
Manual Talairach alignment may be necessary, or
include the -notal-check flag to skip this test,
making sure the -notal-check flag follows -all
or -autorecon1 in the command string.
See http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Talairach