[Insight-developers] DCMTK Questions

Williams, Norman K norman-k-williams at uiowa.edu
Tue May 29 11:55:54 EDT 2012


Jean-Christophe, it seems like what you're saying is that you wish
DCMTK_DIR, if defined, to in essence, force ITK_USE_SYSTEM_DCMTK.  That's
fine, and looking how other external packages are dealt with -- FFTW being
the prime example -- that's how it would work.


Bill, I think that the goal, as indicated by Terry Yu, is that GDCM will
be deprecated, so the DCMTK reader will be the one true way to deal with
DICOM.  This could be a problem for programs that have used GDCMImageIO
explicitly -- or used GDCM calls directly, but I think most people were
cured of that tendency when ITK went from GDCM1 to GDCM2, which broke
anything that used GDCM in a non-trivial fashion.

I think I see my way clear to make DCMTK work as well -- or probably
better -- than GDCM as a DICOM reader.  I need to focus on that and
produce a Gerrit patch.  There are a ton of peripheral issues around
DICOM reading that will take longer to iron out.

The fact of the matter is that DICOM is squirrelly enough to make it
impossible to use transparently with the existing ImageIO framework.  For
one thing, the ImageFileReader/Writer really don't want to deal with
directories as image paths, so any program who wants to use ITK for Image
IO has to have infrastructure around ImageIO to sniff for the DICOM case
and do it differently than every other file format.

Plus DICOM is potentially quite a lot more than just files in a file
system, and ITK has never dealt with the rest of DICOM at all.  Slicer4
has some support for network DICOM stuff, but they're using DCMTK directly
to handle that case.

On 5/29/12 10:28 AM, "Bill Lorensen" <bill.lorensen at gmail.com> wrote:

>If you treat dcmtk as we treat vtk then you will not need a
>ITK_USE_SYSTEM_DCMTK.



________________________________
Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged.  If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited.  Please reply to the sender that you have received the message in error, then delete it.  Thank you.
________________________________


More information about the Insight-developers mailing list