 |
 |
I am using FSL 4.1. Bedpostx does seem to be producing errors for some of the slices. Again, I had to run bedpostx twice, back to back, in order for the merged...samples to be created. For the first run (pid = 2769), there were problems with slices 0000 through 0042. $ cat bedpostx.e2766e-1 ** ERROR (nifti_image_read): failed to find header file for '/home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000' ** ERROR: nifti_image_open(/home/kjyoder/research/conduct_disroder/fs/1001/dmri/data_slice_0000): bad header info ERROR: failed to open file /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000 ERROR: Could not open image /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000 Image Exception : #22 :: Failed to read volume /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000 An exception has be thrown Failed to read volum /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000Trace: read_volume4DROI I confirmed that ~kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0000.nii.gz does exist. It's fslinfo output: data_type FLOAT32 dim1 112 dim2 112 dim3 1 dim4 33 datatype 16 pixdim1 2.0000000000 pixdim2 2.0000000000 pixdim3 2.0000007153 pixdim4 1.0000000000 cal_max 0.0000 cal_min 0.0000 file_type NIFTI-1+ For the second run (pid = 2769, there were problems with slices 0024 through 0041. The seemed to be of two types. First: $ cat bedpostx.e2769e-30 Error: failed to open file /home/kjyoder/research/conduct_disorder/fs/1001/dmri.bedpostX/diff_slices/data_slice_0029/mean_dsamples.nii.gz ERROR: Could not open image /home/kjyoder/research/conduct_disorder/fs/1001/dmri.bedpostX/diff_slices/data_slice_0029/mean_dsamples Image Exception : #23 :: Failed to open volume /home/kjyoder/research/conduct_disorder/fs/1001/dmri.bedpostX/diff_slices/data_slice_0029/mean_dsamples for writing An exception has been thrown Failed to open volume /home/kjyoder/research/conduct_disorder/fs/1001/dmri.bedpostX/diff_slices/data_slice_0029/mean_dsamples for writingTrce: save_basic_volume4D. And another example: ** ERROR (nifti_image_read): short header read '/home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030.nii.gz' ** ERROR: nifti_image_open(/home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030): bad header info ERROR: failed to open file /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030 ERROR: Could not open image /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030 Image Exception : #22 :: Failed to read volume /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030 An exception has been thrown Failed to read volume /home/kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030Trace: read_volume4DROI. ~kjyoder/research/conduct_disorder/fs/1001/dmri/data_slice_0030 does *not* exist, as all of the data_slice_* files have been removed from dmri. I'm starting to wonder if there is a problem with qsub jobs not properly finishing or waiting for prerequisite jobs to finish. Let me know if there's anything else you need. Thanks again, Keith -- Keith Yoder PhD Candidate, Integrative Neuroscience Social Cognitive Neuroscience Laboratory The University of Chicago 5846 S. University Avenue, Kelly 312 Chicago, IL 60637 Phone: 574-215-9678 [log in to unmask] On Fri, Jul 27, 2012 at 12:39 PM, Stamatios Sotiropoulos < [log in to unmask]> wrote: > Hi Keith, > > This is strange. So dtifit ouptut looks fine? For bedpostx are there any > error log files produced for the slices that are not processed? By the way, > which FSL version are you using? > > Stam > > > > On 27 Jul 2012, at 16:21, Keith Yoder wrote: > > I re-ran bedpostx after binarizing the nodif_brain_mask.nii.gz. I've > attached some screenshots of what the volumes look like (the "dti_" is not > on the nifti volume). The commands.txt file is correctly populated with 80 > calls to bedpostx_single_slice.sh, but it seems only some of the slices are > reconstructed into the post-processed volumes. > > data.nii.gz - 112 x 112 x 80 > nodif_brain_mask.nii.gz - 112 x 112 x 80 > > dyads1_dispersion.nii.gz : 112 x 112 x 48 > dyads1.nii.gz : 112 x 112 x 48 > dyads2_dispersion.nii.gz : 112 x 112 x 57 > dyads2.nii.gz : 112 x 112 x 57 > mean_dsamples.nii.gz : 112 x 112 x 66 > mean_f1samples.nii.gz : 112 x 112 x 55 > mean_f2samples.nii.gz : 112 x 112 x 63 > mean_ph1samples.nii.gz : 112 x 112 x 52 > mean_ph2samples.nii.gz : 112 x 112 x 61 > mean_S0samples.nii.gz : 112 x 112 x 67 > mean_th1samples.nii.gz : 112 x 112 x 48 > mean_th2samples.nii.gz : 112 x 112 x 57 > merged_f1samples.nii.gz : 112 x 112 x 55 > merged_f2samples.nii.gz : 112 x 112 x 63 > merged_ph1samples.nii.gz : 112 x 112 x 52 > merged_ph2samples.nii.gz : 112 x 112 x 61 > merged_th1samples.nii.gz : 112 x 112 x 48 > merged_th2samples.nii.gz : 112 x 112 x 57 > > -- > Keith Yoder > PhD Candidate, Integrative Neuroscience > Social Cognitive Neuroscience Laboratory > The University of Chicago > 5846 S. University Avenue, Kelly 312 > Chicago, IL 60637 > Phone: 574-215-9678 > [log in to unmask] > > > > > On Thu, Jul 26, 2012 at 5:52 PM, Stamatios Sotiropoulos < > [log in to unmask]> wrote: > >> Keith, could you please rerun bedpostx with a binary mask? >> >> Cheers >> Stam >> >> >> >> On 26 Jul 2012, at 23:12, Keith Yoder wrote: >> >> Thanks for the quick response. I'm CC'ing the FSL list because it looks >> like bedpostx is not executing properly. >> >> The nodif_brain_mask looks good (I've attached a representative >> screenshot). >> >> However, the bedpostx output is not the right size and does not seem to >> be reconstructing the whole brain. The input image (data.nii.gz) is 112 x >> 112 x 80, but the merged images are not. For instance, >> merged_f1samples.nii.gz is only 112 x 112 x 48. When viewing the volumes, >> most of the inferior slices are simply not there. The attached >> f1_missing_slices.png shows one of these images. >> >> Also strange, the bedpostX directory still has the diff_slices folder, >> but the folder only has data_slice_0016 through data_slice_0028. >> >> -Keith >> -- >> Keith Yoder >> PhD Candidate, Integrative Neuroscience >> Social Cognitive Neuroscience Laboratory >> The University of Chicago >> 5846 S. University Avenue, Kelly 312 >> Chicago, IL 60637 >> Phone: 574-215-9678 >> [log in to unmask] >> >> >> >> On Thu, Jul 26, 2012 at 3:53 PM, Anastasia Yendiki < >> [log in to unmask]> wrote: >> >>> >>> Hi Keith - This seems to be a different problem here, since all the >>> bedpostx outputs are there and are loaded in without a problem. The error >>> occurs after that. What's the size of your images? Does the mask file look >>> normal? >>> >>> Thanks, >>> a.y >>> >>> >>> On Thu, 26 Jul 2012, Keith Yoder wrote: >>> >>> Hi, >>>> >>>> Was there any resolution to this issue? I'm having a similar problem >>>> running trac-all on a Linux cluster. >>>> >>>> I think bedpostx works if I call it directly from the command line >>>> twice. >>>> The first time, bedpostx_postproc.sh doesn't seem to work and the >>>> *samples.nii.gz files are only created if I rerun bedpostx. However, >>>> trac-all...-path fails. I've attached the trac-all-path output and >>>> pasted >>>> the relevant section below: >>>> >>>> Creating output directory/mnt/ide0/home/**kjyoder/research/conduct_** >>>> disorder/fs/1001/dpath/lh.cst_**AS_av >>>> g33_mni_flt >>>> Loading DWIs from >>>> /mnt/ide0/home/kjyoder/**research/conduct_disorder/fs/** >>>> 1001/dmri/dwi.nii.gz >>>> Loading mask from/mnt/ide0/home/kjyoder/**research/conduct_disorder/fs/ >>>> **1001/dlabel/diff/aparc+ >>>> aseg_mask.flt.nii.gz >>>> Loading BEDPOST parameter samples from >>>> /mnt/ide0/home/kjyoder/**research/conduct_disorder/fs/** >>>> 1001/dmri.bedpostX >>>> Loading segmentation map from/mnt/ide0/home/kjyoder/** >>>> research/conduct_disorder/fs/**1001/dlabel/mni/aparc+a >>>> seg.nii.gz >>>> Loading b-values from >>>> /mnt/ide0/home/kjyoder/**research/conduct_disorder/fs/**1001/dmri/bvals >>>> Loading gradients from >>>> /mnt/ide0/home/kjyoder/**research/conduct_disorder/fs/**1001/dmri/bvecs >>>> Segmentation fault (core dumped) >>>> >>>> Here are the contents of the subjects bedpostX directory: >>>> >>>> bvals >>>> bvecs >>>> commands.txt >>>> diff_slices >>>> dyads1_dispersion.nii.gz >>>> dyads1.nii.gz >>>> dyads2_dispersion.nii.gz >>>> dyads2.nii.gz >>>> logs >>>> mean_dsamples.nii.gz >>>> mean_f1samples.nii.gz >>>> mean_f2samples.nii.gz >>>> mean_ph1samples.nii.gz >>>> mean_ph2samples.nii.gz >>>> mean_S0samples.nii.gz >>>> mean_th1samples.nii.gz >>>> mean_th2samples.nii.gz >>>> merged_f1samples.nii.gz >>>> merged_f2samples.nii.gz >>>> merged_ph1samples.nii.gz >>>> merged_ph2samples.nii.gz >>>> merged_th1samples.nii.gz >>>> merged_th2samples.nii.gz >>>> monitor >>>> nodif_brain_mask.nii.gz >>>> xfms >>>> >>>> Also, I'm attaching the bvals, bvecs, dmrirc.local, trac-all.error, and >>>> trac-all.log for the subject. Let me know if there is any other >>>> information >>>> you need. >>>> >>>> Thanks in advance, >>>> Keith >>>> -- >>>> Keith YoderPhD Candidate, Integrative Neuroscience >>>> >>>> Social Cognitive Neuroscience Laboratory >>>> The University of Chicago >>>> 5846 S. University Avenue, Kelly 312 >>>> Chicago, IL 60637 >>>> Phone: 574-215-9678 >>>> [log in to unmask] >>>> >>>> >>>> >>>> >>>> On Fri, Oct 7, 2011 at 10:04 AM, Anastasia Yendiki >>>> <[log in to unmask]> wrote: >>>> >>>> Hi Judit, >>>> >>>> 1. If you can run bedpostx directly then you should do: >>>> bedpostx /usr/local/freesurfer/**subjects/C001/dmri >>>> >>>> Note that this dmri directory and the one you tried to run >>>> bedpostx_single_slice.sh on are different, so I'm not sure which >>>> location is the right one where your data is, but in any case >>>> you don't need to run bedpostx_single_slice.sh. >>>> >>>> 2. The error seems to occur either when it's reading the bvecs >>>> file, or right after that when it's accessing the bedpostx >>>> outputs. So I'd try to look those for that subject and see if >>>> everything is ok. >>>> >>>> Hope this helps, >>>> a.y >>>> >>>> On Fri, 7 Oct 2011, Judit Haasz wrote: >>>> >>>> >>>> Hi, >>>> >>>> I ran into 2 problems related to tracula. >>>> >>>> 1. trac-all -bedp command exits with following >>>> message. >>>> >>>> WARN: Running FSL's bedbost locally - this might >>>> take a while >>>> >> > WARN: It is recommended to run this step on a >>>> cluster >>>> >> > bedpostx_seychelles >>>> /usr/local/freesurfer/**subjects/C001/dmri >>>> >> > subjectdir is >>>> /usr/local/freesurfer/**subjects/C001/dmri >>>> >> > Making bedpostx directory structure >>>> >> > Queuing preprocessing stages >>>> >> > [: 223: NONE: unexpected operator >>>> >> > [: 314: NONE: unexpected operator >>>> >> > [: 327: xbedpostx_pre: unexpected operator >>>> >> > [: 486: x: unexpected operator >>>> >> > [: 486: -le: argument expected >>>> >> > Queuing parallel processing stage >>>> >> > [: 223: NONE: unexpected operator >>>> >> > [: 327: xbedpostx: unexpected operator >>>> >> > [: 486: x53: unexpected operator >>>> >> > 0 slices processed >>>> >> > Queuing post processing stage >>>> >> > [: 223: NONE: unexpected operator >>>> >> > [: 314: NONE: unexpected operator >>>> >> > [: 327: xbedpostx_post: unexpected operator >>>> >> > [: 486: x: unexpected operator >>>> >> > [: 486: -le: argument expected >>>> >> > >>>> >>>> This problem has been posted in May and June. >>>> bedpostx from command line >>>> runs perfectly and that is how I have been running >>>> it so far. One of u >>>> suggested to try the following: >>>> >>>> /usr/local/fsl/bin/bedpostx_**single_slice.sh/disks/open/** >>>> judit/subjects/diffu<http://bedpostx_single_slice.sh/disks/open/judit/subjects/diffu> >>>> sion_tutorial/dtisubjects2005/**subj_523/dmri >>>> 2 1 1000 1250 25 1 0 >>>> >>>> then I got: >>>> >>>> ** ERROR (nifti_image_read): failed to find header >>>> filefor'/disks/open/judit/**subjects/diffusion_tutorial/** >>>> dtisubjects2005/subj_523/ >>>> dmr >>>> i/data_slice_0000' >>>> **ERROR:nifti_image_open(/**disks/open/judit/subjects/** >>>> diffusion_tutorial/dtisub >>>> jects2 >>>> 005/subj_523/dmri/data_slice_**0000): bad header info >>>> ERROR: failed to openfile/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_523/ >>>> dmri >>>> /data_slice_0000 >>>> ERROR: Could not openimage/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_523 >>>> /dmri >>>> /data_slice_0000 >>>> Image Exception : #22 :: Failed to >>>> readvolume/disks/open/judit/**subjects/diffusion_tutorial/** >>>> dtisubjects2005/subj_52 >>>> 3/dmri >>>> /data_slice_0000 >>>> >>>> An exception has been thrown >>>> Failed to readvolume/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_52 >>>> 3/dmri >>>> /data_slice_0000Trace: read_volume4DROI. >>>> >>>> Done >>>> >>>> Can u suggest a reason behind all this? This happens >>>> on Linux machine >>>> (2.6.35-28-generic #50-Ubuntu 10.10, x86_64, Java VM >>>> Version: Java >>>> 1.6.0_17-b04 with Sun Microsystems Inc. Java >>>> HotSpot(TM) 64-Bit Server VM >>>> mixed mode). >>>> On Mac OS X (version 10.7.1 - Lion, 2 x 2.66 GHz >>>> 6-core Intel Xeon >>>> processors, 32 GB 1333 MHz DDR3 RAM) trac-all -bedp >>>> runs well. >>>> >>>> My other question is: >>>> trac-all -path exists with error (in case of a >>>> single subject): >>>> >>>> Loading DWIsfrom/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_501/ >>>> dmri >>>> /dwi.nii.gz >>>> Loading maskfrom/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_501/ >>>> dlab >>>> el/diff/aparc+aseg_mask.bbr.**nii.gz >>>> Loading BEDPOST parameter samplesfrom/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_501/ >>>> dmri >>>> .bedpostX >>>> Loading segmentation mapfrom/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_501/ >>>> dlab >>>> el/mni/aparc+aseg.nii.gz >>>> Loading b-valuesfrom/disks/open/judit/** >>>> subjects/diffusion_tutorial/**dtisubjects2005/subj_501/ >>>> dmri >>>> /bvals >>>> Loading gradientsfrom/disks/open/** >>>> judit/subjects/diffusion_**tutorial/dtisubjects2005/subj_**501/ >>>> dmri >>>> /bvecs >>>> Segmentation fault >>>> Linux Odd 2.6.35-28-generic #50-Ubuntu SMP Fri Mar >>>> 18 18:42:20 UTC 2011 >>>> x86_64 GNU/Linux >>>> >>>> trac-paths exited with ERRORS at Fri Oct 7 10:21:20 >>>> CEST 2011 >>>> >>>> thanks for help. >>>> Judit >>>> >>>> >>>> >>>> >>>> >>>> ______________________________**_________________ >>>> Freesurfer mailing list >>>> [log in to unmask] >>>> https://mail.nmr.mgh.harvard.**edu/mailman/listinfo/**freesurfer<https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer> >>>> >>>> >>>> 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<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. >>>> >>>> >>>> >>>> >>> _______________________________________________ >>> Freesurfer mailing list >>> [log in to unmask] >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >>> >>> 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. >>> >>> >> <nodif_brain_mask.png><f1_missing_slices.png> >> >> >> > <dti_data.png><dti_dtifit_FA.png><dti_mean_dsamples.png> > <dti_nodif_brain_mask.png> > > >