C-STORE-RSP Status - DICOM

This is a discussion on C-STORE-RSP Status - DICOM ; Hello all, What should be the status in C-STORE-RSP if the SOP instance(image) is already existing in Storage SCP. Regards...

+ Reply to Thread
Results 1 to 4 of 4

Thread: C-STORE-RSP Status

  1. C-STORE-RSP Status

    Hello all,
    What should be the status in C-STORE-RSP if the SOP instance(image)
    is already existing in Storage SCP.

    Regards


  2. Receiving duplicate SOP Instances, was Re: C-STORE-RSP Status

    Hi

    Interesting question as to what to do if you already have the
    SOP Instance.

    If it is the "same" as what you already have, no matter, doesn't
    matter what you do, no point in sending a failure status.

    But should one check and what should one check and should one do
    it before sending a response (slow, increases latency and reduces
    throughput) ?

    What if they are not the "same" in some respect ? Should one keep
    the old, or discard it and replace with the new, or coerce (one of
    them) to use new UIDs (maintaining internal consistency in any
    new set of course) ?

    The standard is silent on this, because it doesn't allow for the
    possibility that different instances with the same SOP Instance UID
    can occur.

    Yet they do.

    More questions than answers.

    David

    PS. The short answer is that your response should reflect what you
    do; the question really is what should you do.

    Gurikar wrote:

    > Hello all,
    > What should be the status in C-STORE-RSP if the SOP instance(image)
    > is already existing in Storage SCP.
    >
    > Regards


  3. Re: Receiving duplicate SOP Instances, was Re: C-STORE-RSP Status

    How SCU should interpret C-STORE-RSP. You said, SCP shouldnt send
    failure,so if it sends success, Ideally what SCU should do in this
    scenerio.. Taking that it has(SCU) sent image successfully.

    Regards


  4. Re: Receiving duplicate SOP Instances, was Re: C-STORE-RSP Status

    Gurikar wrote:

    > How SCU should interpret C-STORE-RSP. You said, SCP shouldnt send
    > failure,so if it sends success, Ideally what SCU should do in this
    > scenerio.. Taking that it has(SCU) sent image successfully.


    I don't think that is quite what I said - I said that in the specific
    case where it doesn't matter (the instances are identical and have
    been received twice), then there is no point in sending anything other
    than success. A failure would just be irritating and confusing to the
    user.

    I did not say that one should not send a failure if they really are
    different.

    But does one even check ? And how thoroughly ?

    The standard does not specify what the SCP should do if it receives an
    instance it (supposedly) already has; accordingly, it is not specified
    what response should be sent (success or failure or warning), nor is
    it specified what an SCU should do when it gets any such response
    back.

    Practically speaking, from the SCU perspective, if it gets a success response
    from the SCP, then it can only assume success - whether that means the SCP
    accepted the instance as a replacement or discarded it or segregated or
    coerced it cannot be determined from the response status.

    Likewise, if it gets a failure status, then it cannot assume that this
    is the reason for the failure. The response codes defined in PS 3.4 Annex B
    Table B.2-1 are not specific enough to cover this situation.

    The warning "Coercion of Data Elements" B000 could arguably be used
    if the UIDs were coerced, but this is really intended more for the
    patient and study IDs (as described elsewhere in Annex B), and besides,
    most SCPs are loathe to send warnings, since some SCUs interpret them
    as failures and annoy the user or retry.

    One could argue that more specific response statuses to cover the
    various permutations and combinations could be added to the standard;
    I am not sure if any SCP would be modified to send them or SCU modified
    to interpret them.

    Likewise, what would a user do with a warning or failure message about
    this ?

    David

+ Reply to Thread