Morphometry Statistics

[marianne]

Hi,

I am new in BrainVisa and manage to did everything up to the morphometry statistics.
When i run the morphometry statistic "program", i have an error that i don't understand. I haven't seen any post about that so here it is.
The error is:

Morphometry statistics running
config : C:\DOCUME~1\marianne\LOCALS~1\Temp\bv_78283.cfg
No error

in Morphometry statistics 1 System command exited with non null value : 1

RuntimeError
neuroProcesses.py (1834) in _processExecution:
result = process.execution( self )
morphometry.py (183) in execution:
context.system( progname, tmp.fullPath() )
neuroProcessesGUI.py (1323) in system:
ret = apply( ExecutionContextGUI.system, (self,) + args, kwargs )
neuroProcesses.py (1928) in system:
raise RuntimeError( _t_( 'System command exited with non null value : %s' ) % str( ret ) )

I don't know if it is a problem of memory or something else, but if someone could enlighten me that will be great.
I have another question, i am actually interested by only one sulcus. Is there a way to create a graph with only one sulcus?

Thank you
Marianne

[riviere]

Hi Marianne,

I am afraid we don't have enough information to indentify your problem. The morphometry program seems to have failed, but I don't know if it may be due to incorrect input parameters or to a bug in it...
Could you please send the log file of BrainVisa ? You should run the failing process, then exit BrainVisa, and send the file which should be named something like C:\DOCUME~1\marianne\LOCALS~1\.brainvisa\brainvisa.log. I guess the log window of BV (ctrl+L) displays the log file complete path in the titlebar of its window.

Otherwise, you cannot create a graph with a single sulcus, but once a complete graph is created and identified, you can analyze only one sulcus in the morphometry process if you like: it's the "region" parameter, where you can either type the full label (or labels list) of the sulcus you are interested in, or use the "labels selector" (the "..." button) to graphically select sulci - but this selector is maybe not very intuitive and user friendly... But in any case you need the morphometry program to work.

Denis

[MSh]

Hi Marriane

I also ran into the same problem when working under windows and got exactly the same error message. However, when I tried this with my linux-based Brainvisa version it worked properly.

MSh

[marianne]

Hi Denis,

Sorry for my very very late answer but here is the log file of BrainVisa generated after the error i think.
It is indeed maybe a windows problem, but if you have a fix for that that will be great!
Thanks for your other answer as well!

Thanks
Marianne

[Dominique Geffroy]

Hi Marianne,

The log file you sent is not usable, it missed the log of the morphometry process.
Anyway, I tried the process on a Windows machine and didn't have any error. Are you sure you have correctly filled in the parameters ? Maybe you could send a screen copy of the process and of the log with the error to help me understand the problem.

Dominique

[marianne]

Hi Dominique,

You am not sure i have correctly filled in the parameters, as i am not sure i understand from the help menu (one the side) what and how i am supposed to enter the parameters.
So here is a printscreen of the process the way i tried to run it as well as hopefully the good log of BrainVisa.
What i am trying to do is Morphing statistics, i am actually interested in only one sulcus (STS) that has been labelled in the graph files. I am not sure whether i need to specify that or anything. Or whether the stats file contain stats for all sulcus and that i need to find the one of interest in it.
Any help will be greatly appreciated, even on how to set up the parameters.

Thanks
Marianne

[riviere]

Hi,
Actually, there are some incorrect parameters on the screenshot (I have not looked at the log because I am not under Windows).
- The Region parameter may be used to restrict the morphometric outputs to specific sulci, but in that case you have to follow the exact nomenclature of the sulci: for the STS, right hemisphere, it is: S.T.s._right (case-sensitive, dots and underscores are important, sorry). But you don't really need to use it, then it will ouput measurements for all sulci, it's fast but it will create about 60 files (one per sulcus). The "..." button will allow you to select sulci from the nomenclature and model in a graphical way.
- The output_prefix parameter is the beginning of the full path to output files: so it must contain the directory information, for instance "C:\morphometry\patients_" will lead to files beginning with that, and ending with sulci names: C:\morphometry\patients_S.T.s._right.dat etc. I don't know if it is the cause of your problem.
This output_prefix parameter is not user friendly at all (I can criticise it because I wrote it), I have recently replaced it with a more convenient directory (selectable via a classical file browser) and a file prefix which does not include the directory name, it will be in the next Brainvisa.

Denis

[marianne]

Hi Denis,

So i have tried again with no input in the Region part and the full path in the output_prefix part.
It still does not work, same error.
We identify the STS visually and relabelled some part of it as STS, could it be the reason why it does not work?!
We did not run the automatic sulci recognition (at least i don't think so.)
Could that explain that it does not work?!
If yes, sorry...

Marianne

[Dominique Geffroy]

Hi,

yes, it is probably the reason why it doesn't work. As it is mentionned in the process documentation (http://www.brainvisa.info/doc/brainvisa ... metry.html), the input graphs must be at least partially labelled. So, you should run the automatic sulci recognition and then run the morphometry statistics on the resulting labelled graph.

Dominique

[riviere]

Hi,

For me it explains that you don't get any result, because you did not use the nomenclature of the model which calculates the measurements so the model doesn't find anything to measure. But in my opinion it doesn't explain any error such as the one you get... it should just do nothing, then say proudly "OK I'm done".

Denis

[marianne]

Hey,

Thank you for your answer, i will try again as soon as i manage to run the automatic sulci recognition. I want to mention though that i do have some sulci labelled (about 4), but with custom names, so i am not sure whether it should work or no.

If, as Denis thinks it is not what causes my problem, and if you have any ideas on what this/these problem(s) might be and on how to fix it/them, i'll take them and try...

Merci
Marianne