DICOM PS3.4 2024b - Service Class Specifications

FF.2.4 Additional Volumetric Considerations

FF.2.4.1 Annotations in Volumetric Presentations States

The Volumetric Presentation States provide two ways for annotating views:

  • Annotations on the Volumetric Presentation View

  • Annotations described by coordinates in the Volumetric Presentation State Reference Coordinate System (VPS-RCS) with optional references to Structured Reports providing context.

Annotations on the view provide the application of free unformatted text or vector graphics as described in the Section C.10.5 “Graphic Annotation Module” in PS3.3. Since the Graphic Annotation Module allows only the addition of graphics to the 2D view defined by the Presentation State without attached clinical meaning, Volumetric Graphic Annotations provide a mechanism to create annotations in the VPS-RCS with optional references to other objects which can have structured context attached.

Volumetric Graphic Annotations can be specified in two variants: either via Graphic Types with 3D coordinates, as defined in Section C. “Graphic Type” in PS3.3, or via a reference to inputs of the Presentation State. The latter is intended to be used to display annotation labels for segmentations of the volume data; for example, when a lesion has been marked via a Segmentation IOD and this segmentation is rendered together with the anatomical data.

Since annotations which are added via the Graphic Annotation Module are defined within the display space, they should not be used to point to clinical relevant structures which would be positioned on a different anatomy after manipulation.

In contrast since Volumetric Graphic Annotations have coordinates in the VPS-RCS, applications can still show them after a user has manipulated the initial view which has been defined by the Presentation State.

The exact visual representation of the annotations is at the discretion of the display application, as well as the mechanisms which may be employed to ensure that Volumetric Graphic Annotations are sufficiently visible, even if the location in the volume is not visible in the current view. E.g. for a Graphic Type POINT a display application might render a crosshair at the specified position in the volume or a sphere with an arrow pointing to it instead of rendering Volumetric Graphic Annotations directly within the volume a projection of the annotations may be rendered as an overlay on top of the view.

However, annotations can be grouped into Graphic Layers and it is suggested that applications provide mechanisms to define rendering styles per Graphic Layer.

See Section XXX.3.4 “Replacing Set of Derived Images With Single VPS Using Crosscurve Animation” in PS3.17 and Section XXX.3.5 “Volumetric Annotations (example: Trajectory Planning)” in PS3.17 for examples of Volumetric Graphic Annotations.

DICOM PS3.4 2024b - Service Class Specifications