[Insight-users] Fwd: Help on DICOM image reading bug

[Mathieu Malaterre]

S Sam wrote:
> Since my first try to send this message was bounced back. I have to 
> resend it again and remove the DICOM image due to restricted bytes in 
> the message.
>  
> If you are interested in solving this problem, please let me know your 
> e-mail address, and I will send the DICOM image immediately.
>  
> Thank you again.
> Sam
> 
> Dear all ITK users and developers;
> 
>  
> 
> I am encountering a problem in reading DICOM image and I am attaching 
> the image asking for your help and advice.
> 
>  
> 
> Since the pixel type of the DICOM image is "14U", I modified the file 
> itkGDCMImageIO.cxx accordingly and recompiled and generated the ITK 
> library. After getting the bufferpointer using GetBufferPointer, it was 
> found that about half of the image data were lost, and the image display 
> using OpenGL indicates obviously that upper part of the image is blank. 
> I tested the code using a TIFF image whose dimension is larger than that 
> of the DICOM image, it works very well. That leads to me doubt about if 
> there is a bug in GDCMImageIO for reading such DICOM image or something 
> wrong with my code. The following is the major part of the code:
> 

Sam,

	Thanks a lot for sending the sample DICOM dataset. I can reproduce the 
problem. Unfortunately right now ITK-GDCM does not handle the reading of 
such dataset. It will require some change in the gdcm layer.

	Please add a bug report at:

	http://itk.org/Bug

	Simply copy/paste your email and assign the bug to me.

Thanks again for the bug report,
Mathieu