Anonymiaztion, Tag 0008,1250.

[Salim Kanoun]

Hi There,

I wanted some information about the anonymization of 0008,1250 tag.

This type 3 Tag store the seriesUID related to the current dicom
https://dicom.innolitics.com/ciods/cr-image/general-series/00081250

Problem, while this tag is a unique key to link an original DICOM, it is not included in the anonymization table defined by the Nema Guideline
http://dicom.nema.org/medical/dicom/current/output/html/part15.html#table_E.1-1

I would want to know what would be the recommendation for this tag, either Clear or Replace (with the new generated UID).

Is there someone from Nema able to say if this tag will be included in the anonymization list ?

Best regards,

Salim

[David Rodriguez]

Hi Salim,

In the DICOM standard part 15 that you mention you can find in section E.1.1 De-Identifier:
"The Attributes listed in Table E.1-1 for each profile are contained in Standard IODs, or may be contained in Standard Extended IODs. An implementation claiming conformance to an Application Level Confidentiality Profile as a de-identifier shall protect or retain all instances of the Attributes listed in Table E.1-1, whether contained in the main dataset or embedded in an Item of a Sequence of Items. "

So you need to apply the actions appropriate for each of the embedded attributes, like Study Instance UID or Series Instance UID in this case.

Notice also the following in the same section: "Some profiles do not allow selective protection of parts of a Sequence of Items. If an Attribute to be protected is contained in a Sequence of Items, the complete Sequence of Items may need to be protected."

Best regards,

David

[Salim Kanoun]

Thanks for you answer, so I understand that we should map the new UIDs this sequences.
I'm going to see with Orthanc if it is technically feasible, I will require to know what SeriesUID is going to be affected at the beginning of the anonymization process.

Best regards,

Salim

[guiot...@gmail.com]

One trick we used to avoid keeping track of this mapping is to hash the source uid with the same algorithm and key. and prefix the result with a valid dicom root.
This way one input uid always gives the same output uid. So if one uid is referred to in multiple instances and multiple tags, all cross references are kept.

KR