Problem with Tools->spm->segment/normalize (using VBM)

Questions about BrainVisa usage and installation

Moderators: denghien, riviere

Post Reply
leka0024
Posts: 19
Joined: Wed Jul 06, 2016 9:52 pm

Problem with Tools->spm->segment/normalize (using VBM)

Post by leka0024 »

I open this process, select the subject from the BV database (already imported the .nii file for the subject), select the parameters, then click run.

The window says the script finished running in about 14 seconds.

In the BV database under the subject, the following new directories and files are created:

- BV_database/analyzes/ (dir -- supposed to have an analysis output here in .../VBMSegmentation/, but this sub-folder isn't created)
- BV_database/subjects/sub_A/t1mri/VBMSegmentation/ (dir -- some outputs are supposed to be here)
- BV_database/subjects/sub_A/t1mri/VBMSegmentation/segmentation/ (dir -- some outputs are supposed to be here)

But the folders are empty!! Where are the smoothed/modulated .nii file outputs that can be used for VBM analysis??

The process window shows the file names that are supposed to be created, but they aren't there even though the process doesn't report any errors.


(Note 1: BV requires the Matlab version of SPM for this process

Note 2: I did see in SPM manual that a cmdline setting in spm_defaults.m has to be set to 'true', and I did change that

Note 3: I have already run this subject through the Morphologist 2015 (non-UI) pipeline, if that matters. But since new directories were created, I think it doesn't matter)
leka0024
Posts: 19
Joined: Wed Jul 06, 2016 9:52 pm

Re: Problem with Tools->spm->segment/normalize (using VBM)

Post by leka0024 »

Hello,

Can you please comment on this?

Thank you!
User avatar
riviere
Site Admin
Posts: 1361
Joined: Tue Jan 06, 2004 12:21 pm
Location: CEA NeuroSpin, Saint Aubin, France
Contact:

Re: Problem with Tools->spm->segment/normalize (using VBM)

Post by riviere »

Hi,
Sorry for the delay, I was on vacation...
I have trying the process, and for me it ran for far longer (10 minutes), so I guess in your case it just did not run SPM at all, and the process does probably not detect failures properly.
The first thing is to check SPM config in brainvisa (preferences, matlab and SPM tabs).
Apparently the process currently only supports SPM8 (not SPM12) so it could maybe the cause of the problem.

You can also check the log in brainvisa (brainvisa/show log menu, or ctrl-L). After running the process it shoud display an item "segment/normalize (using VBM toolbox)", which you can open, it should contain sub-items, one for the "no-link" variant of the process, and inside it, one for the matlab execution. Mine starts like this:

Code: Select all

/neurospin/local/bin/matlab 
Command line
'/neurospin/local/bin/matlab' '-nosplash' '-nodisplay' '-r' 'vbm8_bv_13031_7'
Output

                            < M A T L A B (R) >
                  Copyright 1984-2015 The MathWorks, Inc.
                   R2015a (8.5.0.197613) 64-bit (glnxa64)
                             February 12, 2015


 

To get started, type one of these: helpwin, helpdesk, or demo.

For product information, visit www.mathworks.com.

 

%--------------------------------------------------------------------%

SPM2         spm2
SPM5         spm5
SPM8         spm8
SPM12        spm12
                            http://www.fil.ion.ucl.ac.uk/spm/software/
%--------------------------------------------------------------------%
FieldTrip    fieldtrip
                                        http://fieldtrip.fcdonders.nl/
%--------------------------------------------------------------------%
BrainStorm   brainstorm
                                 http://neuroimage.usc.edu/brainstorm/
%--------------------------------------------------------------------%
EEGLAB       eeglab
                                          http://sccn.ucsd.edu/eeglab/
%--------------------------------------------------------------------%
MILRIM       milrim
                 https://bioproj.extra.cea.fr/redmine/projects/milrim/
%--------------------------------------------------------------------%

 ___  ____  __  __                                            

/ __)(  _ \(  \/  )                                           
\__ \ )___/ )    (   Statistical Parametric Mapping           

(___/(__)  (_/\/\_)  SPM8 - http://www.fil.ion.ucl.ac.uk/spm/




Initialising SPM.
.
.
Duplicate application tag in applications 'SPM' and 'SPM'.

Duplicate application tag in applications 'BasicIO' and 'BasicIO'.

.
.
.
.
.
.
done


SPM present working directory:
	/tmp



------------------------------------------------------------------------
Running job #1
------------------------------------------------------------------------

Running 'VBM8: Estimate & Write'

Initial Coarse Affine Registration..

Fine Affine Registration..

VBM8 r435: /tmp/bv_13031_7.nii

NLM-Filter with multi-threading

Using 8 processors

Skull-stripping using graph-cut

Amap with Kmeans

Nu correction







Nu correction


MRF priors: alpha 0.182 0.434 0.384 	 beta 1.265
and ends with:

Code: Select all

------------------------------------------------------------------------
Running job #2
------------------------------------------------------------------------

Running 'Print'


Printing to
/tmp/spm_2016Aug22.ps

Done    'Print'

Done


Done    'VBM8: Estimate & Write'

Done


Bye for now...


Result
Value returned = 0
I guess you don't have all this, and even not the SPM startup message ?

Denis
leka0024
Posts: 19
Joined: Wed Jul 06, 2016 9:52 pm

Re: Problem with Tools->spm->segment/normalize (using VBM)

Post by leka0024 »

My turn to say sorry for the delay!

I am now leaving this project (it was just a summer project for me), but want to say thanks a lot for all the support! Possibly new members on the project could post here in the future, seeking support.


As to the problem:
I didn't realize VBM was a separate toolbox for SPM! I thought it was supposed to use SPM natively.

In fact, once I realized this error I had a lot of success with VBM analysis using SPM12 (Matlab version) and the newest version of the VBM toolbox for SPM12 called "CAT". http://www.neuro.uni-jena.de/cat/index.html
It's quite easy to use and seems very powerful.

Anyone looking to do VBM analysis, I would highly suggest giving this a try.
Post Reply