[Ctk-developers] Fwd: [GitHub] modalitiesInStudy not read correctly and more Modality related issues [commontk/CTK GH-15]

[Jean-Christophe Fillion-Robin]

Is somebody looking into that issue ?

Thanks
Jc

---------- Forwarded message ----------
From: toromand <
reply+i-819199-208fc263772333bb5a07781cdaa80466812e9900 at reply.github.com>
Date: Wed, Apr 27, 2011 at 7:29 AM
Subject: [GitHub] modalitiesInStudy not read correctly and more Modality
related issues [commontk/CTK GH-15]
To: jchris.fillionr at kitware.com


1.
in CTK/Libs/DICOM/Core/ctkDICOMDatabase.cpp
the tag DCM_ModalitiesInStudy is not read correctly
Instead of:
dataset->findAndGetOFString(DCM_ModalitiesInStudy, modalitiesInStudy);
it should be:
dataset->findAndGetOFStringArray(DCM_ModalitiesInStudy, modalitiesInStudy);
to allow reading multiple modalities (XX\YY\ZZ)

2.
Also, you might consider more "standard" way of performing the query with
the modalities filter. Although the fact that many SCP implement search by
the ModalitiesInStudy tag, it is not according to the DICOM standard and is
rather a IHE extension with undefined behavior when used as a search
criterion (well, it works well with dcm4chee). So it might now work with all
DICOM compliant SCPs. The "proper" way (according to DICOM specifications)
would be to perform a query for all needed modalities and combine the
results.

3. Rather more important issue is that the current database design lacks a
VERY important thing. In DICOM, modalities are part of Series, not the
Study, and the ModalitiesInStudy is "informative" and new feature that you
should use on Study level for information on the Study, but in your current
implementation, you don't have the information on the actual modality of
every series. This is a very important information and it is missing
although it exists in the dataset. So at present with (1) not fixed you get
for study with 2 modalities (XA/OT)
STUDY (XA)
 - SERIES 1
 - SERIES 2
instead of
STUDY (XA/OT)  -----> *with (1) fixed
 - SERIES 1 (XA)
 - SERIES 2 (OT)

Regards
Dragan Toroman

--
Reply to this email directly or view it on GitHub:
https://github.com/commontk/CTK/issues/15



-- 
+1 919 869 8849
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/ctk-developers/attachments/20110502/77452d5d/attachment.htm>