hello,

does dicom offer an end of study message? that is, how can i tell at my review station that i will not be receiving any more images for a patient study? i do not want to review a patient until i know all the acquired images have been received, so it would be great to get an indication that the study is complete before i review the case.

cheers,

joel

The IHE profile scheduled workflow specifies which DICOM and HL7 messages should be used to support such a use case.
More information can be found at: https://wiki.ihe.net/index.php/Scheduled_Workflow

On Monday, May 17, 2021 at 5:48:18 PM UTC-5, feldman_joel@yahoo.com wrote:
> hello,
>
> does dicom offer an end of study message? that is, how can i tell at my review station that i will not be receiving any more images for a patient study? i do not want to review a patient until i know all the acquired images have been received, so it would be great to get an indication that the study is complete before i review the case.
>
> cheers,
>
> joel

As specified by the referenced IHE profile, you could use DICOM MPPS, but that is sparsely used, although widely deployed and poorly implemented as some modalities send it prior to having all the images sent to the PACS. Most PACS systems rely on operator input to close out the study as they typically do some kind of QA anyway which result in a HL7 order "complete" transaction and triggers both elimination from the modality worklist and adding to the radiology worklist. Some use IAN as well (Instance Availability).

Short version: You need an externally managed reading work list.

A DICOM Study is never finished ... at any point in time in the future additional series can be added (re-read, addendum, post-processing, retrospective reconstructions).

Two common questions arise operationally:

- is a Study "ready" for something (e.g., "ready to read", one example of triggering next workflow step)?
- is a Study "finished" wrt. billing (i.e., "ready to bill", perhaps when report is complete and signed)

Consider, for example, a study sent elsewhere for a referral, where it may be re-read (and the read re-billed); it is still the same DICOM Study (same Study Instance UID, usually not coerced on ingestion).

Obviously, existing systems may impose their own constraints (e.g., lock the study so that it can be read; may or may not require manual intervention to re-open to add more), but that is not a DICOM constraint.

DICOM does not provide a "Study Complete" flag nor any service/operation to "complete" a Study (since Detached Study Management and Study Content Notification were retired).

Do not confuse a Procedure Step (Study Component) with a Study. IHE SWF addresses the former, not the latter.

A workstation receiving C-STORE operations passively (only) cannot reliably know when it has "enough" (or all that is available) - this needs to be "managed" by external means (e.g., with a reading worklist, whose state changes are handled by out of band transactions, e.g., operator pushing the "ready to read" button in the RIS or other workflow manager +/- rules about what else is required, e.g., CAD results present).

Note that a C-FIND will tell you what the other end (currently) has (released), but that is not necessarily everything coming/needed (e.g., consider CAD or 3D lab work product that may be delayed).

David