Invalid sop instance uid - bad Dicom file

[bkasmai]

The Dicom documentation(https://dicom.nema.org/dicom/2013/outpu ... ter_B.html) states that:

Because of the flexibility allowed by the DICOM Standard in creating organizationally derived UIDs, implementations should not depend on any assumed structure of UIDs and should not attempt to parse UIDs to extract the semantics of some of its components."

I have anonymised a study and changed its sop instance uid using a private SOP Instance UID for re-export to PACS for teaching purposes. I get the following error:


dcmsend -v -r -aet MYSCPU -aec PACSSCP 10.145.92.63 4006 test.dcm
I: checking input files ...
E: cannot add DICOM file to the transfer list: test.dcm: Invalid SOP Instance UID
F: bad DICOM file: test.dcm: Invalid SOP Instance UID

The study could be loaded to a few known Dicom browsers with no plroblem.

Any help on this would be appreciated.

[Michael Onken]

Hi,

can you post the generated UID? e.g. run dcmdump +P SOPInstanceUID <yourfile.dcm>

Best regards,
Michael

[bkasmai]

dcmdump +P SOPInstanceUID test.dcm
(0008,0018) UI [1.2.826.0.1.3680043.2.1012.00.637.30.08.2022.11.45.50] # 54, 1 SOPInstanceUID

Thanks for looking into this.

[J. Riesmeier]

The ".00." in the UID is not allowed. It violates the definition of the "Unique Identifier" (UI) Value Representation.

[bkasmai]

Thank you! Mystery solved!

[bkasmai]

I am puzzled. My applcation only changes Study and Series Instance UID. I think the SOPInstanceUId is changed by Pydicom package which is supposed to be the bread and butter of medical image datasets!

[bkasmai]

Sorry, ignore my last message. Nothing to do with pydicom, I am a lousy programmer!

[bkasmai]

Sorry to reopen this again. I changed the program to avoid .00. but still the same problem
dcmdump +P SOPInstanceUID test.dcm
(0008,0018) UI [1.2.826.0.1.3680043.2.1012.0.542.06.09.2022.10.50.16] # 52, 1 SOPInstanceUID

[J. Riesmeier]

Leading zeros ("0") are also not allowed, e.g. in ".06.09.". See UID Encoding Rules in DICOM PS3.5.

[bkasmai]

So did i misunderstand the quote below from Dicom documentation?

implementations should not depend on any assumed structure of UIDs and should not attempt to parse UIDs to extract the semantics of some of its components

[Marco Eichelberg]

That quote only means that applications are not supposed to read meaning into the structure of the UIDs (e.g. they must not assume that UID 1.2.3.4 is in any way related to UID 1.2.3). That does not prevent an application to check whether or not a UID is syntactically valid, i.e. contains no leading zeroes, is not longer than 64 characters, uses a zero byte instead of a space character as a pad byte if necessary.

[bkasmai]

Thank you for your help. This was educational for me.