DICOM PS3.17 2019a - Explanatory Information

GG JPIP Referenced Pixel Data Transfer Syntax Negotiation (Informative)

The JPIP Referenced Pixel Data transfer syntaxes allow transfer of image objects with a reference to a non-DICOM network service that provides the pixel data rather than encoding the pixel data in (7FE0,0010).

The use cases for this extension to the Standard relate to an application's desire to gain access to a portion of DICOM pixel data without the need to wait for reception of all the pixel data. Examples are:

  1. Stack Navigation of a large CT Study.

    In this case, it is desirable to quickly scroll through this large set of data at a lower resolution and once the anatomy of interest is located the full resolution data is presented. Initially lower resolution images are requested from the server for the purpose of stack navigation. Once a specific image is identified the system requests the rest of the detail from the server.

  2. Large Single Image Navigation.

    In cases such as microscopy, very large images may be generated. It is undesirable to wait for the complete pixel data to be loaded when only a small portion of the specific image is of interest. Additionally, this large image may exceed the display capabilities thus resulting in a decimation of the image when displayed. A lower resolution image (i.e., one that matches the resolution of the display) is all that is required, as additional data cannot be fully rendered. Once an area of interest is determined, the application can pan and zoom to this area and request additional detail to fill the screen resolution.

  3. Thumbnails.

    It is desirable to generate thumbnail representations for a study. This has been accomplished through various means, many of which require the client to receive the complete pixel data from the server to generate the thumbnail image. This uses significant network bandwidth.

    The thumbnails can be considered low-resolution representations of the image. The application can request a low-resolution representation of the image for use as a thumbnail.

  4. Display by Dimension.

    Multi-frame Images may encode multiple dimensions. It is desirable for an application to access only the specific frames of interest in a particular dimension without the need to receive the complete pixel data. By using the multi-dimensional description, applications using the JPIP protocol may request frames of the Multi-frame Image.

The association negotiation between the initiator and acceptor controls when this method of transfer is used. An acceptor can potentially accept both the JPIP Referenced Pixel Data transfer syntax and a non-JPIP transfer syntax on different presentation contexts. When an acceptor accepts both of these transfer syntaxes, the initiator chooses the presentation context.

Examples:

For the following cases:

Case 1:

Case 2:

For the following cases:

Case 3:

Case 4:

DICOM PS3.17 2019a - Explanatory Information