[Insight-users] re:re:why not ITK can read the spacing correctly?

[Mathieu Malaterre]

Yixun Liu wrote:
> Hi Luis,
> Thank your help!
> Yes I compare it with the dicom header.
>  
> One question. If the dicom files are not consecutive. Could ITK read 
> them correctly?

No. Special care was done in GDCM so that if non-consecutive -let say
scout views- are provided as input ITK-GDCM will let the user know that
a 3D volume can not be reconstructed from the input images.


> Another question. I use DICOMImageIO2 and DicomImageIO to read a same 
> series. The DICOMImageIO2 can read  the spacing[2] correctly, but 
> DICOMImageIO is error. Why?

As other people mention, you should not use DICOM2 anymore. If you use
Dicom you should be fine since it transparently default to calling
GDCMImageIO. In anyway -I insist- the spacing calculation done in
ITK-GDCM is done correctly.

We really appreciate user feedback, but in your case you do not provide
a lot of information on your conclusion, technically this is also known
as FUD(*) ;)
So, please, spend the time to provide any/all of the following:
* What other software you are using,
* Some slices from your data,
* a dcmdump/DicomPrintPatientInformation of your data,
* Tcl/Python/c++ code to reproduce the bug.

Please keep in mind that user discover ITK, in browsing this mailing
list, so if you believe you found a bug, please provide a detailed
explanation of the issue, a way to reproduce or any details that
*undoubtly* demonstrate the issue in ITK.

Regards
Mathieu
(*)http://en.wikipedia.org/wiki/FUD