Hi...

I am doing shape analysis for hipocampi.But I got these two errors:

1)in ParaToSPHARMMesh
 
BestFlip:4,3
vnl_error_matrix_dimension:vnl_tag_mul: Dimensions [3,4002] and [1002,3] do not match.
Aborted

2) in GenParaMesh

Euler Number ok = 2
flag = 0 iterations performed: 194
flag = 0 iterations performed: 197
initially active: {}.
fromC[expr_] := expr /. (mant_Real)*e + (expon_Integer) :> mant*10^expon;


How can I fix the problem?

Thanks
Gozde

Hi Godze,


About error 1) it seems you are using a parametric field (all the information contained in the Phi.txt Theta.txt PhiHalf.txt) and SPHARM meshes with a different number of subdivisions. 1002 number of points is associated with 10 subdivisions, 4002 is associated with 20 subdivisions. It seems rather weird that that happens if you ran the command from the scratch, all parametric fields and meshes are generated with the same number of points, does it make sense? Did you run ParaToSPHARMMesh before? Can you try to re-run the command by erasing all the outputs from ParaToSPHARMMesh?

About error 2), does that happens after a long time computing, or just in the begining? it seems to me that this is normal screen messages for the program. Could you provide a little bit more information on it?

HTH
Thanks,

BEATRIZ

Hi Beatriz
 
 
Thanks for your response.
 
About error 2), that happens after a long time computing so I ended the program
 
 
Thanks,
 
 
Gozde

Hi Godze,


That sounds like a numerical error in GenParaMeshCLP. I would revise the input segmentation (remove holes, hole-like rims and very thin areas), remove all the associated outputs from SegPostProcess and re-run.

HTH

BEATRIZ

Hi Beatriz
 
Thanks for your response.
 
First I didn't use CLP version. Today I tried CLP version but  I got these error in SegPostProcessCLP:
 
WARNING: In /biomed-resimg/crews_rodent/devel/linux/ITK_3.X/Insight_3.20.0/InsightToolkit-3.20.0/Code/IO/itkGDCMImageIO.cxx, line 348
GDCMImageIO (0x89c7fa8): The DICOM file: 1L.gipl does not have a preamble.


itk::ExceptionObject (0xbf9eed88)
Location: "void itk::GDCMImageIO::InternalReadImageInformation(std::ifstream&)"
File: /biomed-resimg/crews_rodent/devel/linux/ITK_3.X/Insight_3.20.0/InsightToolkit-3.20.0/Code/IO/itkGDCMImageIO.cxx
Line: 805
Description: itk::ERROR: GDCMImageIO(0x89c7fa8): Could not read header from file: 1.gipl
Reason: Success
 
How can I fix the problem? Do I have to draw all images again with itk-snap?

Hi Godze,


When you open 1.gipl and 1L.gipl with itk-snap, do you have any problems?

BEATRIZ