Hi Bruce,
 
I found on http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Talairach
 

Fixing Bad Output From the Talairach Registration

FreeSurfer computes a linear Talairach transform contained in a 3x4 matrix in a file called talairach.xfm and is located in the <subject name>/mri/transforms directory. Under some circumstances, the alignment can fail the automatic failure detection scheme (-tal-check flag, which can be disabled via -notal-check).

Indeed, when I run recon-all with -notal-check, I donot see:
 
*#@# Talairach Failure Detection Wed Apr  6 12:36:53 EDT 2011
in recon-all-staus.log file any more.
 
Does this mean I have to do the manual talairach registration every time I run my recon-all, as it is a time consuming process?
Is there any thing (from an acquisition perspective or otherwise) to eliminate this problem?
 
 
Thanks in advance.
 


 
On Thu, Apr 7, 2011 at 12:27 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote:
but you don't know what else was running on the same node at the time, do you? Can you lock the amount of memory you will need? I think to be safe you want 3G, but maybe Nick can chime in


On Thu, 7 Apr 2011, Saad Ramadan wrote:

Hi Bruce,

it is a cluster here a Brigham and wouldnot expect it to be low on memory.
When I run a shell:
free -mt

I get:


                total       used       free     shared    buffers
cached
Mem:          3950       2918       1031          0        163       1596
-/+ buffers/cache:       1158       2791
Swap:         3820        125       3694
Total:          7770       3044       4725


On Thu, Apr 7, 2011 at 10:52 AM, Bruce Fischl <fischl@nmr.mgh.harvard.edu>wrote:

how much memory do you have in that machine?

Bruce

On Thu, 7 Apr 2011, Saad Ramadan wrote:

Dear all,


I ran (on an isotropic MPRAGE ):

recon-all -subject  mysubject  -i
./DICOM/test.25.1.2010.04.27.21.51.18.31250.14004670.IMA -autorecon-all

which ran for few hours and then exited without a successful end as can be
seen on log file below.


Contents of *recon-all-status.log* are:

Wed Apr  6 12:31:55 EDT 2011
#@# MotionCor Wed Apr  6 12:33:10 EDT 2011
#@# Nu Intensity Correction Wed Apr  6 12:33:40 EDT 2011
#@# Talairach Wed Apr  6 12:35:22 EDT 2011
*#@# Talairach Failure Detection Wed Apr  6 12:36:53 EDT 2011
*#@# Intensity Normalization Wed Apr  6 12:36:53 EDT 2011
#@# Skull Stripping Wed Apr  6 12:41:44 EDT 2011
#@# EM Registration Wed Apr  6 14:16:04 EDT 2011
#@# CA Normalize Wed Apr  6 15:18:54 EDT 2011
#@# CA Reg Wed Apr  6 15:22:59 EDT 2011

The last few lines of *recon-all.log*   produces:

0091: dt=0.112000, rms=0.700 (-0.001%), neg=0, invalid=766
resetting metric properties...
setting smoothness coefficient to 10.00
blurring input image with Gaussian with sigma=2.000...
0000: dt=0.000, rms=0.638, neg=0, invalid=766
0092: dt=0.772947, rms=0.623 (2.335%), neg=0, invalid=766
0093: dt=0.028000, rms=0.622 (0.145%), neg=0, invalid=766
0094: dt=0.028000, rms=0.622 (-0.063%), neg=0, invalid=766
blurring input image with Gaussian with sigma=0.500...
0000: dt=0.000, rms=0.622, neg=0, invalid=766


Do you have any idea what is going wrong here?

Thanks



 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.