search

SIMEXP / niak

GNU Octave library for connectivity analysis in large fMRI datasets
http://niak.simexp-lab.org
Other
28 stars 21 forks source link

nu_correct error #368

Closed nooralsh closed 6 years ago

nooralsh commented 7 years ago

I'm processing some fMRI data using the singularity distribution of NIAK_cog-1.1.3 and I am consistently running into an error at the nu_correct stage for the anatomical T1 volume. This may be similar to issue #366 but I will provide some more detail.

The first odd thing is that even though my script is set as follows:

opt.t1_preprocess.nu_correct.arg = '-distance 75';

you can see below that the -distance flag is set to 200. Following that is the error itself which I don't know how to address. I also could not find a way to turn off the nu_correct to bypass this issue (which I was able to do with the nu_correct for fMRI which was also failing). Any help would be appreciated, thanks!

******************************
Log of the (octave) job : t1_preprocess_sp0107
Started on 19-Sep-2017 20:11:13
User: jvogel44
host : sw-4r01-n25
system : unix
******************************
***********************************
Preprocessing of a T1 brain volume
***********************************
Original brain volume : /gs/project/eim-670-aa/nooralsh/PING_niak/p0107/anat/anat.nii.gz

*****************************************
Non-uniformity correction on an MR volume
*****************************************
Copying or converting file /gs/project/eim-670-aa/nooralsh/PING_niak/p0107/anat/anat.nii.gz to /loc
alscratch/99691332.gm-1r16-n04.guillimin.clumeq.ca/niak_tmp_t1_preprocess_sp0107_435576994_anat/vol
.mnc
Copying or converting file gb_niak_omitted to /localscratch/99691332.gm-1r16-n04.guillimin.clumeq.c
a/niak_tmp_t1_preprocess_sp0107_435576994_anat/mask.mnc
Running NU_CORRECT with the following command:
nu_correct -clobber -tmpdir /localscratch/99691332.gm-1r16-n04.guillimin.clumeq.ca/niak_tmp_t1_prep
rocess_sp0107_435576994_anat/ -distance 200 /localscratch/99691332.gm-1r16-n04.guillimin.clumeq.ca/
niak_tmp_t1_preprocess_sp0107_435576994_anat/vol.mnc /localscratch/99691332.gm-1r16-n04.guillimin.c
lumeq.ca/niak_tmp_t1_preprocess_sp0107_435576994_anat/vol_nu.mnc

Can't use 'defined(@array)' (Maybe you should just omit the defined()?) at /gs/project/gsf-624-aa/q
uarantaine/minc-toolkit-0.3.18/bin/nu_estimate_np_and_em line 165.
nu_correct: crashed while running nu_estimate_np_and_em (termination status=65280)
ans =  255
Copying or converting file /localscratch/99691332.gm-1r16-n04.guillimin.clumeq.ca/niak_tmp_t1_prepr
ocess_sp0107_435576994_anat/vol_nu.mnc to /gs/project/eim-670-aa/nooralsh/PING_niak/preproc/fmri_pr
eprocess_allniak_cog-1.1.3_2017-09-19/anat/sp0107/anat_sp0107_nuc_nativet1.nii.gz

Something went bad ... the job has FAILED !
The last error message occured was :
Couldn't find any file fitting the description /localscratch/99691332.gm-1r16-n04.guillimin.clumeq.
ca/niak_tmp_t1_preprocess_sp0107_435576994_anat/vol_nu.mnc
File /usr/local/niak/commands/read_write/niak_read_vol.m at line 188
File /usr/local/niak/commands/read_write/niak_cp_fmri.m at line 79
File /usr/local/niak/bricks/misc/niak_brick_copy.m at line 147
File /usr/local/niak/bricks/t1_processing/niak_brick_nu_correct.m at line 222
File /usr/local/niak/bricks/t1_processing/niak_brick_t1_preprocess.m at line 494
File /usr/local/niak/extensions/psom-2.3.1/psom_run_job.m at line 185
File /usr/local/niak/extensions/psom-2.3.1/psom_run_job.m at line 106
File /usr/local/niak/extensions/psom-2.3.1/psom_worker.m at line 161

Here is the output from the worker.oqsub file:

----------------------------------------
execution ['/bin/bash', '/gs/project/eim-670-aa/nooralsh/PING_niak/preproc/fmri_preprocess_allniak_cog-1.1.3_2017-09-19/logs/tmp/psom1.sh']
psom1 done with with returncode 0
touch /gs/project/eim-670-aa/nooralsh/PING_niak/preproc/fmri_preprocess_allniak_cog-1.1.3_2017-09-19/logs/worker/psom1/worker.failed
touch /gs/project/eim-670-aa/nooralsh/PING_niak/preproc/fmri_preprocess_allniak_cog-1.1.3_2017-09-19/logs/worker/psom1/worker.exit
touch /gs/project/eim-670-aa/nooralsh/PING_niak/preproc/fmri_preprocess_allniak_cog-1.1.3_2017-09-19/logs/worker/psom1/worker.out
all workers are done with their jobs
----------------------------------------
Begin PBS Epilogue Tue Sep 19 17:19:03 EDT 2017 1505855943
Job ID:     99691332.gm-1r16-n04.guillimin.clumeq.ca
Username:   jvogel44
Group:      eim-670-01
Job Name:   psom1
Session:    13535
Limits:     neednodes=1:ppn=1:sandybridge,nodes=1:ppn=1:sandybridge,pmem=2700mb,walltime=48:00:00
Resources:  cput=01:04:13,energy_used=0,mem=1203660kb,vmem=1628244kb,walltime=01:11:06
Queue:      sw-serial
Account:    eim-670-aa
Exit code:  0
Nodes:  sw-4r01-n25
Killing leftovers...

End PBS Epilogue Tue Sep 19 17:19:04 EDT 2017 1505855944
----------------------------------------
pbellec commented 7 years ago

Thanks a lot for the report.

The issue #366 is for nu-correct of nii fmri. It is a known bug that needs fixing.

The anat nu correct should work in nifti though and I am not familiar with the error message. It may have to do with that particular release/container. @poquirion any suggestion?