error compute cortical fold graph (3.1)

[joost]

Hi,

i am using version 3.1.

when using "compute cortical fold graph", L/R*.arg are created but with every brain i get the following error during the process:

output:
Process Compute Cortical Fold Graph 6 started on 2008/06/13 11:25
Masking Bias corrected image with hemisphere masks...
Computing skeleton and buried gyrus watershed...
Building Attributed Relational Graph...

A.I.M.S. assertion failed: d < D, file /home/a-sac-ns-research/build-stable-Fedora-4-i686-release/include/aims/vector/vector.h, line 571
terminate called after throwing an instance of 'carto::assert_error'
what(): d < D

RuntimeError
neuroProcesses.py (2090) in _processExecution: result = process.execution( self )
CorticalFoldsGraphThickness.py (62) in execution:
'-w', self.white_mesh, '-l', self.hemi_mesh, '-o', self.output_graph )
neuroProcessesGUI.py (1260) in system:
ret = apply( ExecutionContextGUI.system, (self,) + args, kwargs )
neuroProcesses.py (2172) in system:
ret = self._system( command, self._systemStdout, self._systemStderr )
neuroProcesses.py (2230) in _system:
result = c.wait()
neuroProcesses.py (1111) in wait:
raise RuntimeError( _t_( 'System call exited abnormally' ) )

is this a problem? processing the same brain in version 3.02 does NOT give me any error.

merci

-joost

[riviere]

Hi,
Hm, right, there is a bug here. I'll fix it...
But it should only appear in the newer variant of the cortical folds graph process, which is not selected by default.
The older one should work as in 3.0.2.
Denis

[riviere]

OK I found it.
We'll make another binary package next week, but we're waiting to fix a few other bugs (a build problem on Mac that prevents running on Intel machines, especially).
By then, you can patch your 3.1 binary package, by copying this file to the corresponding location in the package, starting from 'python' in the package root dir (overwrite the existing one):
ftp://ftp.cea.fr/pub/dsv/anatomist/upda ... _init__.py.

Denis

[joost]

Hi,

i´m using the latest version of Brainvisa (3.1 - 2008/06/20). when using compute cortical fold graph i _still_ get the following error message (an *.arg file with info is created though):

*****
Process Compute Cortical Fold Graph 1 started on 2008/06/24 10:16
Masking Bias corrected image with hemisphere masks...
Computing skeleton and buried gyrus watershed...
Building Attributed Relational Graph...

in Cortical Fold Graph Thickness and Volumes 1 System call exited abnormally
RuntimeError
neuroProcesses.py (2090) in _processExecution: result = process.execution( self )
CorticalFoldsGraphThickness.py (62) in execution:
?
neuroProcessesGUI.py (1260) in system:
ret = apply( ExecutionContextGUI.system, (self,) + args, kwargs )
neuroProcesses.py (2172) in system:
ret = self._system( command, self._systemStdout, self._systemStderr )
neuroProcesses.py (2230) in _system:
result = c.wait()
neuroProcesses.py (1111) in wait:
raise RuntimeError( _t_( 'System call exited abnormally' ) )

in Compute Cortical Fold Graph 1 System call exited abnormally
RuntimeError
neuroProcesses.py (2090) in _processExecution: result = process.execution( self )
CorticalFoldsGraph.py (125) in execution:
?
neuroProcesses.py (1584) in run:
return self._run( context )
neuroProcesses.py (1698) in _run:
return context.runProcess( self._process )
neuroProcesses.py (1943) in runProcess:
result = self._processExecution( _process, None )
neuroProcesses.py (2090) in _processExecution:
result = process.execution( self )
CorticalFoldsGraphThickness.py (62) in execution:
?
neuroProcessesGUI.py (1260) in system:
ret = apply( ExecutionContextGUI.system, (self,) + args, kwargs )
neuroProcesses.py (2172) in system:
ret = self._system( command, self._systemStdout, self._systemStderr )
neuroProcesses.py (2230) in _system:
result = c.wait()
neuroProcesses.py (1111) in wait:
raise RuntimeError( _t_( 'System call exited abnormally' ) )
************

merci,
-joost

[Dominique Geffroy]

Hi,

I cannot reproduce this error. Which package did you download (linux, windows, mac) ?

Maybe the problem is linked to the data. Have you checked the previous steps' results ?

Dominique

[joost]

problem solved. thanks!

a question on the resulting L/R*.arg files. they contain a "brain_hull_area" and "refbrain_hull_area". what are the precise definitions of these? for example, what isotropic closing-distance was used to ensure the smoothness of the boundary?

merci,
-joost

[riviere]

Hi,
The "brain hull" is the closed hull: the brain without folds. Its area is more or less the area of the extern part of the gyri. You can visualize it on the skeleton image for instance.
All "ref*" attributes are calculated in the Talairach space (after AC/PC and scale factors transformation).
Denis

[Jean-Francois Mangin]

Once again Arnaud should know which parameter wasa used, I try to get in touch with him.