Mirc CTP anonymizer

  • I have data sets that are passed through Mirc CTP anonymizer (http://mircwiki.rsna.org/index.php?title=MIRC_CTP) end up in conquest server. These data sets have PatientID removed intentionally. Both the CTP and conquest severs are only stages in the anonymization pipeline. Anomymized data gets sent out to Horos PACS.


    I use conquest instead of DCMTK's storescp since the storescp tends to create multiple directories for the same StudyInstanceUID. Data sets are CT images in thousands.


    Problem is that the Conquest server re-introduces PatientID tag in data sets with 0000000 as value in them. Since I just want to use conquest server to receive the data sets, and not to store permanent, I have to do one extra step of removing these PatientID tags from data sets from several thousands of images, using DCMTK's dcmodify. I use FileNameSyntax as %name\%sopuid.dcm in dicom.ini.


    Perhaps have PatientID as mandatory is the default behavior of any PACS. I am just wondering it would be ever possible not to introduce PatientID on data sets, when it is not present in data sets. Please let me know. Thanks.

  • CTP's dicom export sends data directly to conquest's AET & port.


    >>The safest fix would be to set patient ID 0000000 to empy there
    Conquest does this automatically when it sees data sets with no PatientID. And, I like I mentioned before, I would rather not see PatientID tag in anonymized data sets. We already anonymize PatientName.


    If PatientID tag is mandatory in conquest, I can simply copy PatientName value into PatientID tag as work around.

  • Hi,


    you can do that on import. patientID is a primary key, it is mandatory.


    try add this to dicom.ini:


    [lua]
    ImportConverter0 = if Data.PatientID=='' then Data.PatientID=Data.PatientName end


    Marcel

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!