• Trending in TID 1500

    From Antje Schroeder@21:1/5 to All on Mon Oct 19 23:33:32 2020
    Good morning.

    We are looking into encoding trending information in a DICOM SR (TID 1500) using actual date and time for each measurement and not just a label, e.g. to document change of a nodule size over the course of time.

    During our discussion we found the following two options:
    * TID 321 offers the means to add a TCOORD item to a measurement, however looking into TID 321 it seems to be thought to be used in conjunction with waveforms, rather than providing  a single time stamp
    * TID 1502 allows to give a Time Point Context, however that only assigns labels not an actual DATETIME to a measurement, which we would like to see in order allow correct treanding

    Is there a best practice standard conformant way to achieve this, or is that something that should be added to TID 1500 and subordinate templates?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From =?UTF-8?Q?J=C3=B6rg_Riesmeier?=@21:1/5 to All on Tue Oct 20 00:28:35 2020
    We are looking into encoding trending information in a DICOM SR (TID 1500) using actual date and time for each measurement and not just a label, e.g. to document change of a nodule size over the course of time.

    What about using Observation​ DateTime​ (0040,A032)​ from the Document Relationship Macro, which allows for specifying "the date and time on which [this] a Content Item was completed": http://dicom.nema.org/medical/dicom/current/output/chtml/part03/
    sect_C.17.3.html#table_C.17-6

    Regards,
    Jörg

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Antje Schroeder@21:1/5 to All on Tue Oct 20 01:21:29 2020
    Jörg Riesmeier schrieb am Dienstag, 20. Oktober 2020 um 09:28:38 UTC+2:
    We are looking into encoding trending information in a DICOM SR (TID 1500) using actual date and time for each measurement and not just a label, e.g. to document change of a nodule size over the course of time.
    What about using Observation​ DateTime​ (0040,A032)​ from the Document Relationship Macro, which allows for specifying "the date and time on which [this] a Content Item was completed": http://dicom.nema.org/medical/dicom/current/output/chtml/
    part03/sect_C.17.3.html#table_C.17-6

    Regards,
    Jörg

    Thanks.
    We thought about that as well, but unfortunately the time point when AI is run (and therefore, when the Observation DateTime is set) does not necessarily match the same time, when the images/study was taken. However for trending the time when the images
    were taken, is more relevant.
    E.g you might run AI on prior exams right around the time when running AI on the current exam. So the Observation DateTime would be roughly the same, but to see how the finding changed over the course of time, it is important to know when the prior study
    was acquired.
    Best regards
    Antje

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From =?UTF-8?Q?J=C3=B6rg_Riesmeier?=@21:1/5 to All on Tue Oct 20 12:33:53 2020
    Dear Antje,

    However for trending the time when the images were taken, is more relevant.

    so, you mean the date/time when the images were acquired? What about Acquisition Date/Time in TID 1602 (Image Library Entry Descriptors) then? See: http://dicom.nema.org/medical/dicom/current/output/chtml/part16/chapter_A.html#sect_TID_1602

    Regards,
    Jörg

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)