Re: [Freesurfer] FreeSurfer recon-all two sets of error

The process is still running but nothing new written in the last 24 hrs.

Excerpts from recon-all.log

......

77 defects to be corrected 0 vertices coincident reading input surface /home/#########/subjects/./surf/lh.qsphere.nofix... reading brain volume from brain... reading wm segmentation from wm...

.......

CORRECTING DEFECT 20 (vertices=71, convex hull=55) After retessellation of defect 20, euler #=-3 (58272,166451,108176) : difference with theory (-54) = -51

CORRECTING DEFECT 21 (vertices=17, convex hull=34) After retessellation of defect 21, euler #=-2 (58273,166465,108190) : difference with theory (-53) = -51

CORRECTING DEFECT 22 (vertices=39536, convex hull=7215)

Help!

On Wed, Feb 27, 2013 at 11:37 AM, Blessy M blessam@gmail.com wrote:

1. It is not contrast enhanced.

2. I am rerunning recon-all after giving it a manually skull stripped

image. Hopefully this will work.

So basically did this:

recon-all -subjid . -autorecon1 -noskullstrip -notal-check cp T1.mgz brainmask.auto.mgz ln -s brainmask.auto.mgz brainmask.mgz recon-all -subjid . -autorecon2 -autorecon3

On Tue, Feb 26, 2013 at 3:10 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

...

thanks. I actually meant a tksurfer image, but this might do. Two questions:

1. Is this a contrast-enhanced acquisition? That will make things more

difficult, although we have processed them in the past.

2. I think the skull stripping removed some brain tissue, which may be

the source of your large defects. Can you please check it and get back to us? This also might be due to the contrast enhancement.

cheers

Bruce

On Tue, 26 Feb 2013, Blessy M wrote:

I should have send it with an extension. But it is tiff, just as you

...

asked.

On Tue, Feb 26, 2013 at 3:03 PM, Bruce Fischl < fischl@nmr.mgh.harvard.edu> wrote: Hi Blessy

  what format is this? If you don't send it with an extension there

is no way to know.... Bruce On Tue, 26 Feb 2013, Blessy M wrote:

        Forgot the image. please find attached.


        On Tue, Feb 26, 2013 at 2:45 PM, Blessy M <blessam@gmail.com>

wrote: I ran this command, and got the attached image tkmedit ./ brainmask.mgz lh.inflated.nofix

              For some reason tksurfer is not looking good/complete
              tksurfer ./ lh inflated


              On Tue, Feb 26, 2013 at 11:53 AM, Bruce Fischl <

fischl@nmr.mgh.harvard.edu> wrote: you can save a tif of it and post it to the list On Tue, 26 Feb 2013, Blessy M wrote:

                          Yes it is still at defect 32.

                          The tail of the recon-all.log is as

follows:

                          CORRECTING DEFECT 30 (vertices=18, convex

hull=22) After retessellation of defect 30, euler #=-7 (59491,172395,112897) : difference with theory (-28) = -21

                          CORRECTING DEFECT 31 (vertices=244, convex

hull=142) After retessellation of defect 31, euler #=-6 (59502,172495,112987) : difference with theory (-27) = -21

                          CORRECTING DEFECT 32 (vertices=44865,

convex hull=8969)

                          Should I attach the lh.inflated.nofix in a

separate email to you, Bruce?

                          On Tue, Feb 26, 2013 at 10:08 AM, Bruce

Fischl fischl@nmr.mgh.harvard.edu wrote: is it still on defect 32? What does the tail of the recon-all.log say? And can you send us an image of the lh.inflated.nofix (if it is the lh that is running)?

                                On Tue, 26 Feb 2013, Blessy M wrote:

                                      I verified lh.orig.nofix, and

it looks fine. But, there has been no new files created since Saturday (2/23/2013).

                                      I am considering stopping the

process.

                                      I was thinking, I had created

a skull stripped image, and was giving that as input to recon-all in the command below: recon-all -subjid . -noskullstrip -autorecon1 -notal-check -autorecon2 -autorecon3

                                      Should -noskullstrip flag be

after autorecon1?

                                      On Mon, Feb 25, 2013 at 3:35

PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: try looking at the lh.inflated.nofix or lh.orig.nofix (or rh, whichever one is running) and see if something is dramatically wrong (like skull attached to brain, or hemis connected, etc....) On Mon, 25 Feb 2013, Blessy M wrote:

                                                  Recon is still

running. I started the process on 2/21/2013. Its been 4 days.

                                                  I ran this

command, and viewed the aseg file, and that looks reasonable tkmedit ./ brainmask.mgz -aux T1.mgz -surfs -aseg

                                                  Currently it is at

this stage:

                                                  Correction of the

Topology Finding true center and radius of Spherical Surface...done Surface centered at (0,0,0) with radius 100.0 in 13 iterations marking ambiguous vertices... 124766 ambiguous faces found in tessellation segmenting defects... ......

                                                  61 defects to be

corrected 0 vertices coincident ...... ...... CORRECTING DEFECT 31 (vertices=244, convex hull=142) After retessellation of defect 31, euler #=-6 (59502,172495,112987) : difference with theory (-27) = -21

                                                  CORRECTING DEFECT

32 (vertices=44865, convex hull=8969)

                                                  On Fri, Feb 22,

2013 at 5:33 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: I don't think either one of these are errors, just warnings that occur pretty frequently and I don't think should impact the results. Does the recon finish? Do the results look ok?

                                                        On Fri, 22

Feb 2013, Blessy M wrote:

                                                              I am

getting the bottom two sets of errors while doing a simple recon-all.

                                                              More

specifically while running this command:

recon-all -subjid . -noskullstrip -autorecon1 -notal-check -autorecon2

-autorecon3

                                                              Has

someone encountered this kind of errors? Is there a fix?

                                                              1)

Computing MAP estimate using 2772 samples...

---

                                                              IFLAG=

-1 LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH ERROR RETURN OF LINE SEARCH: INFO= 6 POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE

INCORRECT OR INCORRECT TOLERANCESoutof QuasiNewtonEMA: 011: -log(p) =

7296.5 tol 0.000010 ... ...

                                                              2)

unfolding failed - restoring original position

                                                              0146:

dt=13.779160, rms=0.814 (0.000%), neg=0, invalid=766

blurring input image with Gaussian with sigma=0.500... 0000: dt=0.000, rms=0.814, neg=0, invalid=766

gcamFindOptimalTimeStep: Complete in 101542.133 ms iter 0, gcam->neg = 801

......................

unfolding failed - restoring original position

                                                              0158:

dt=13.828393, rms=0.814 (-0.069%), neg=0, invalid=766

blurring input image with Gaussian with sigma=0.500... 0000: dt=0.000, rms=0.814, neg=0, invalid=766

gcamFindOptimalTimeStep: Complete in 100610.148 ms

                                                  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/**compliancelinehttp://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.