Specify StudyDescription when creating DICOMSeg ?

[salimkanoun]

Hi there, thanks for this very helpfull library,

Correct me if i'm wrong but I didn't find anyway to add StudyDescription to a generated DICOMSeg,

In highdicom.seg.Segmentation and highdicom.SOPClass there is not attribut for StudyDescription,

It is normal or just a forgotten dicom tag ?

Best regards,

Salim

[JoaoSantinha]

Hi @salimkanoun ,

If I am not mistaken it should have the same study description of the original study description of the DICOM series for which the segmentation was created. Unfortunately, I am not able to test it right now, but hopefully it will help. Additionally, that tag is optional and study UID should allow for the segmentation to be under the right study

[salimkanoun]

Dear @JoaoSantinha thanks for your answer !
I'm using copy_patient_and_study_information but the resulting dicom seg does not have the study description tag

[CPBridge]

Hi @salimkanoun, @JoaoSantinha is correct, the "StudyDescription" is copied from the source images because it should match the "StudyDescription" in every other file in the study (being a study-level attribute). If you are not seeing it in the segmentation, you should find that that is because it is missing in the source files, thus it is "correct" to omit it from the segmentation.

If you really need to override this (though I wouldn't recommend it because StudyDescription should be consistent throughout the study), you can do this using the pydicom-level API and simply set it after construction (seg.StudyDescription = "My Study").

If you check and find that study description is present in the source images but missing from the Segmentation, then there is indeed a bug in the library that we should fix.

[CPBridge]

Also, you should not call the copy_patient_and_study_information directly, this is done for you by the Segmentation constructor. I actually want to deprecate that method at some point, it was a mistake to make it public in my opinion. I think I'll add something to the docstring to make this clearer

[CPBridge]

I think I'll add something to the docstring to make this clearer

See #385

[salimkanoun]

Dear all, thank you so much for you support,

I redone some unit testing in my code, you're correct the StudyDescription is well copied from original study, sorry for the false positive report