Relative X-Ray Exposure (0018,1405) wrong VR? - DICOM

This is a discussion on Relative X-Ray Exposure (0018,1405) wrong VR? - DICOM ; Hi all, we have a situation with a DR scanner from Imaging Dynamics Company Ltd. The client is requesting that we show in the viewer what they see as F# in the console. Right after the F# they see a ...

+ Reply to Thread
Results 1 to 2 of 2

Thread: Relative X-Ray Exposure (0018,1405) wrong VR?

  1. Relative X-Ray Exposure (0018,1405) wrong VR?

    Hi all,

    we have a situation with a DR scanner from Imaging Dynamics Company
    Ltd.

    The client is requesting that we show in the viewer what they see as F#
    in the console. Right after the F# they see a number, this number is a
    negative double (-0.51 for instance). Our client told us it was the
    Relative Xray Exposure but we already have code handling that tag
    (0018,1405). Using dcmdump from offis it says:

    (0018,1405) IS [-0.51] # 6, 1
    RelativeXRayExposure

    However our viewer was showing 0 and not -0.51. Then we looked up
    RelativeXRayExposure in DICOM and we found this:

    (0018,1405) Relative X-ray Exposure IS 1

    And from part 5 we see the definition for an IS:

    """
    Integer String

    A string of characters representing an Integer
    in base-10 (decimal), shall contain only the
    characters 0 - 9, with an optional leading "+" or
    "-". It may be padded with leading and/or
    trailing spaces. Embedded spaces are not
    allowed.
    The integer, n, represented shall be in the
    range:
    -231 <= n <= (231 - 1).

    Valid values:
    "0"-"9", "+", "-" of
    Default Character
    Repertoire

    Size:
    12 bytes
    maximum
    """

    So, this is an integer, a -0.51 can't be stored in there. Our code,
    using another library than offis' is converting that to 0. We contacted
    the vendor to see what they suggest and they say they can put the value
    in a private tag 4321,0005. That does not help us much, but we're
    getting around to getting this "resolved".

    However, my question here is about DICOM, the definition of the
    RelativeXRayExposure from DICOM says:

    """
    Indication of the applied dose, in
    manufacturer specific units.
    Notes: 1. This value is intended to provide
    a single location where
    manufacturer specific information
    can be found for annotation on a
    display or film, that has meaning to
    a knowledgeable observer.
    2. This may be a calculated or
    measured value. Examples are the
    detector entrance dose (KB), the
    CR sensitivity value (S), or the
    logarithmic median (lgM).
    """

    >From that description it sounds like the VR should really be DS and not

    IS.

    I see there was an old post about this DICOM tag:

    http://groups.google.com/group/comp....2?dmode=source

    Does anybody know what are we supposed to do in this situation? The DR
    vendor is trying to store something in a field that its definition
    sounds good, but the VR chosen does not look to be the good one.

    Any comments?

    -Anibal


  2. Re: Relative X-Ray Exposure (0018,1405) wrong VR? [Imaging DynamicsDR]

    Hi Anibal

    It is indeed irritating that (0018,1405) is an IS, but it has
    been that way for a very long time (it was in the 1993 original
    DICOM standard).

    If the acquisition modality is sending decimal values in that
    attribute, then they are creating illegal objects, and they
    should stop. I notice that they do specifically mention this
    deliberate violation in their DICOM Conformance Statement on
    their web site, but that is no excuse for creating invalid
    objects (one should not be providing configuration options
    to allow creation of invalid objects).

    Since there is no standard definition for how to interpret the
    units of values in this attribute, it would be better if the
    vendor sent larger numbers, perhaps scaled by x1000, e.g. -510,
    instead of -0.51. That way they could convey the same information
    and everyone could display it without violating the standard (and
    continue to send their private tag if they want as well).

    Hopefully, AAPM TG#116 will soon define a standard meaning for
    this sort of concept, and an appropriate DICOM attribute will
    be added or the meaning of the existing one defined.

    David

    Anibal Jodorcovsky wrote:
    > Hi all,
    >
    > we have a situation with a DR scanner from Imaging Dynamics Company
    > Ltd.
    >
    > The client is requesting that we show in the viewer what they see as F#
    > in the console. Right after the F# they see a number, this number is a
    > negative double (-0.51 for instance). Our client told us it was the
    > Relative Xray Exposure but we already have code handling that tag
    > (0018,1405). Using dcmdump from offis it says:
    >
    > (0018,1405) IS [-0.51] # 6, 1
    > RelativeXRayExposure
    >
    > However our viewer was showing 0 and not -0.51. Then we looked up
    > RelativeXRayExposure in DICOM and we found this:
    >
    > (0018,1405) Relative X-ray Exposure IS 1
    >
    > And from part 5 we see the definition for an IS:
    >
    > """
    > Integer String
    >
    > A string of characters representing an Integer
    > in base-10 (decimal), shall contain only the
    > characters 0 - 9, with an optional leading "+" or
    > "-". It may be padded with leading and/or
    > trailing spaces. Embedded spaces are not
    > allowed.
    > The integer, n, represented shall be in the
    > range:
    > -231 <= n <= (231 - 1).
    >
    > Valid values:
    > "0"-"9", "+", "-" of
    > Default Character
    > Repertoire
    >
    > Size:
    > 12 bytes
    > maximum
    > """
    >
    > So, this is an integer, a -0.51 can't be stored in there. Our code,
    > using another library than offis' is converting that to 0. We contacted
    > the vendor to see what they suggest and they say they can put the value
    > in a private tag 4321,0005. That does not help us much, but we're
    > getting around to getting this "resolved".
    >
    > However, my question here is about DICOM, the definition of the
    > RelativeXRayExposure from DICOM says:
    >
    > """
    > Indication of the applied dose, in
    > manufacturer specific units.
    > Notes: 1. This value is intended to provide
    > a single location where
    > manufacturer specific information
    > can be found for annotation on a
    > display or film, that has meaning to
    > a knowledgeable observer.
    > 2. This may be a calculated or
    > measured value. Examples are the
    > detector entrance dose (KB), the
    > CR sensitivity value (S), or the
    > logarithmic median (lgM).
    > """
    >
    >>From that description it sounds like the VR should really be DS and not

    > IS.
    >
    > I see there was an old post about this DICOM tag:
    >
    > http://groups.google.com/group/comp....2?dmode=source
    >
    > Does anybody know what are we supposed to do in this situation? The DR
    > vendor is trying to store something in a field that its definition
    > sounds good, but the VR chosen does not look to be the good one.
    >
    > Any comments?
    >
    > -Anibal
    >


+ Reply to Thread