Understanding how to display DICOM RT-Dose - DICOM

This is a discussion on Understanding how to display DICOM RT-Dose - DICOM ; Hi all, I am trying to better understand the concepts behind DICOM RT-Dose objects, and how they should be displayed by viewing applications supporting DICOM RT objects. What I understood until now (correct me if I am wrong) is that ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: Understanding how to display DICOM RT-Dose

  1. Understanding how to display DICOM RT-Dose

    Hi all,

    I am trying to better understand the concepts behind DICOM RT-Dose
    objects, and how they should be displayed by viewing applications
    supporting DICOM RT objects.

    What I understood until now (correct me if I am wrong) is that doses
    can be expressed in terms of:

    1) dose points or isodose curves: in this case, the elements which
    should be displayed by a RT viewer application are described in the
    DICOM dataset through the ROI Contour module, which basically
    describes points, lines or polygons as a set of 3D coordinates in the
    3D space of the patient.

    2) does grid: in this second case, the dose values are described
    through the Pixel Data data element. Here, pixel values actually
    represent dose values, correct? A multi-frame Pixel Data attribute may
    be present, and in this case each frame represent dose values for a
    given "plane / slice", as specified by the "(300C,000C) - Grid Frame
    Offset Vector" data element.

    Is my understanding correct until now?

    Ok, now I have a question:

    I have some sample studies where multiple "RT-Dose" DICOM files are
    present, each one having multiple frames in Pixel Data. Considering
    all these RT-Dose DICOM files, I have several RT dose "pixel data"
    frames (coming from different files) which apply to a given CT slice
    (i.e., which belong to the same plane as the CT slice). Should I SUM
    the dose values coming from each RT dose frame which apply to a given
    CT slice in order to obtain the actual dose value which should be
    applied to every pixel of my CT slice? Or how should I combine values
    coming from different RT dose frames which apply to the same CT slice
    (i.e., which have the same Grid Frame Offset value)?

    In particular, I am considering the issue from the point of view of
    the viewing application: I have seen applications which show the RT
    dose which applies to each CT slice as a semi-transparent overlay area
    varying from blue for lower-dose zones to red for higher-dose zones
    (IsoShades). So in this case I guess the viewing application should
    compute the "overall" dose which applies to a given CT slice. How are
    the several RT-dose frames related to a given CT slice's plane
    combined?

    Any clarification would be greatly appreciated.

    Regards,

    John.


  2. Re: Understanding how to display DICOM RT-Dose

    Hi John,
    Let me add in my views....
    Dose can contains the Grid doses which stroed in pixeldata (ImagPixel
    Module Attribute) , you have to multiply this each pixel value with
    DoseGridScaling(0x3004000E) then you will retrieves the Griddoses
    which is in dose units specified in DoseUnits attributes.

    IsoDoseCurves or IsoDosePoints can be stored in ROICOntour and
    structureset , roi dose module points of curves which having same dose
    across the curves....

    To view this dose , you have to get the current CT image plane cut
    this plane in dose volume which constructed based on pixeldata and
    dose grid scaling display the overlay on ctimage.. At a time one dose
    only you have to display on top of ct volume..
    Regards,
    KK


    johnwhit...@gmail.com wrote:
    > Hi all,
    >
    > I am trying to better understand the concepts behind DICOM RT-Dose
    > objects, and how they should be displayed by viewing applications
    > supporting DICOM RT objects.
    >
    > What I understood until now (correct me if I am wrong) is that doses
    > can be expressed in terms of:
    >
    > 1) dose points or isodose curves: in this case, the elements which
    > should be displayed by a RT viewer application are described in the
    > DICOM dataset through the ROI Contour module, which basically
    > describes points, lines or polygons as a set of 3D coordinates in the
    > 3D space of the patient.
    >
    > 2) does grid: in this second case, the dose values are described
    > through the Pixel Data data element. Here, pixel values actually
    > represent dose values, correct? A multi-frame Pixel Data attribute may
    > be present, and in this case each frame represent dose values for a
    > given "plane / slice", as specified by the "(300C,000C) - Grid Frame
    > Offset Vector" data element.
    >
    > Is my understanding correct until now?
    >
    > Ok, now I have a question:
    >
    > I have some sample studies where multiple "RT-Dose" DICOM files are
    > present, each one having multiple frames in Pixel Data. Considering
    > all these RT-Dose DICOM files, I have several RT dose "pixel data"
    > frames (coming from different files) which apply to a given CT slice
    > (i.e., which belong to the same plane as the CT slice). Should I SUM
    > the dose values coming from each RT dose frame which apply to a given
    > CT slice in order to obtain the actual dose value which should be
    > applied to every pixel of my CT slice? Or how should I combine values
    > coming from different RT dose frames which apply to the same CT slice
    > (i.e., which have the same Grid Frame Offset value)?
    >
    > In particular, I am considering the issue from the point of view of
    > the viewing application: I have seen applications which show the RT
    > dose which applies to each CT slice as a semi-transparent overlay area
    > varying from blue for lower-dose zones to red for higher-dose zones
    > (IsoShades). So in this case I guess the viewing application should
    > compute the "overall" dose which applies to a given CT slice. How are
    > the several RT-dose frames related to a given CT slice's plane
    > combined?
    >
    > Any clarification would be greatly appreciated.
    >
    > Regards,
    >
    > John.


  3. Re: Understanding how to display DICOM RT-Dose

    John,
    An RT-Dose represents a calculated dose. It references an RT-Plan,
    which, in turn, references an RT-Structure Set, which, in turn,
    references a set of images (usually CT's). Calculation of Dose is
    based primarily on the RT-Plan, but may involve use of data in the RT-
    Structure Set, and even the CT's. It also involves models of
    Radiation Source which are outside of DICOM.

    The Dose, the RT-Structure Set, and the CT's are almost always in the
    same frame of reference, so (for display purposes) you can shortcut
    all of these references if you know that the Dose and the CT's are in
    the same frame of reference. The Dose represents the calculated dose
    for a given "summation type" and a particular plan. A summation type
    can be "PLAN", "FRACTION", "BEAM", "BRACHY", or "CONTROL_POINT" (In my
    experience, I have only seen "PLAN", "FRACTION", and "BEAM"). A
    "PLAN" is the most inclusive summation type. (Currently an RT-Dose
    cannot represent a dose for more than a single plan).

    Most often the summation type is "PLAN", i.e. the RT-Dose represents
    the Dose for the entire plan. It is wrong to sum the doses for
    multiple plans. Summing the doses for multiple beams in the same
    plan, on the other hand makes more sense.

    I don't know the particular characteristics of the RT-Dose and CT's
    that you have, but the planes of grid based RT-Dose are often not co-
    incident with the CT's upon which you would display the dose, so you
    often have to interpolate between the dose planes.

    Generally, questions of dose summation are best addressed to a medical
    physicist, not a DICOM expert, as the really relevant questions relate
    the the underlying physics, and not to DICOM encoding of dose.

    HTH,
    Bill B


    On Mar 14, 1:45*pm, johnwhit...@gmail.com wrote:
    > Hi all,
    >
    > I am trying to better understand the concepts behind DICOM RT-Dose
    > objects, and how they should be displayed by viewing applications
    > supporting DICOM RT objects.
    >
    > What I understood until now (correct me if I am wrong) is that doses
    > can be expressed in terms of:
    >
    > 1) dose points or isodose curves: in this case, the elements which
    > should be displayed by a RT viewer application are described in the
    > DICOM dataset through the ROI Contour module, which basically
    > describes points, lines or polygons as a set of 3D coordinates in the
    > 3D space of the patient.
    >
    > 2) does grid: in this second case, the dose values are described
    > through the Pixel Data data element. Here, pixel values actually
    > represent dose values, correct? A multi-frame Pixel Data attribute may
    > be present, and in this case each frame represent dose values for a
    > given "plane / slice", as specified by the "(300C,000C) - Grid Frame
    > Offset Vector" data element.
    >
    > Is my understanding correct until now?
    >
    > Ok, now I have a question:
    >
    > I have some sample studies where multiple "RT-Dose" DICOM files are
    > present, each one having multiple frames in Pixel Data. Considering
    > all these RT-Dose DICOM files, I have several RT dose "pixel data"
    > frames (coming from different files) which apply to a given CT slice
    > (i.e., which belong to the same plane as the CT slice). Should I SUM
    > the dose values coming from each RT dose frame which apply to a given
    > CT slice in order to obtain the actual dose value which should be
    > applied to every pixel of my CT slice? Or how should I combine values
    > coming from different RT dose frames which apply to the same CT slice
    > (i.e., which have the same Grid Frame Offset value)?
    >
    > In particular, I am considering the issue from the point of view of
    > the viewing application: I have seen applications which show the RT
    > dose which applies to each CT slice as a semi-transparent overlay area
    > varying from blue for lower-dose zones to red for higher-dose zones
    > (IsoShades). So in this case I guess the viewing application should
    > compute the "overall" dose which applies to a given CT slice. How are
    > the several RT-dose frames related to a given CT slice's plane
    > combined?
    >
    > Any clarification would be greatly appreciated.
    >
    > Regards,
    >
    > John.



  4. Re: Understanding how to display DICOM RT-Dose

    Thanks to everybody for your precious replies.

    The RT-Dose DICOM files I am dealing with have "summation type" =
    "BEAM", and there are multiple RT-Dose files (i.e., multiple RT-Dose
    grid pixel data) referring to a given CT plane (image), and referring
    to a single RT-Plan DICOM object. It looks like there is one RT-Dose
    file for each beam described in the RT-Plan.

    In this case, do you believe it would be appropriate to sum all RT-
    Dose grid pixel values (after having multiplied them by the
    "DoseGridScaling" value) related to each beam in order to build up the
    "overall" dose image to be displayed as an overlay to the CT image?

    Finally, a more general question: can you recommend a documentation
    source where I could find some more information about RT? I agree with
    you that some of my questions are more "medical physicist's" questions
    than DICOM-related questions, but I don't know where to find
    appropriate information.
    For instance, now I would also like to understand something more about
    how to display the various beams described by the RT-Plan DICOM object
    over the CT image(s) they refer to, and I guess I would need some
    background information about RT Plan and how the radiation beams are
    setup in order to better understand the meaning of the various values
    stored in the RT Plan DICOM object...

    Any suggestion would be highly appreciated.

    Best regards,

    John.


+ Reply to Thread