Page tree
Skip to end of metadata
Go to start of metadata

The FHIR Management Group requested that II attempt to extend the DocumentReference resource to support the ImagingReference use cases. One approach would be to add the additional fields to the Context section of that resource

Extension to content element

This is a revised draft based on feedback from John Moehrke.

In this approach:

  • attachment fields are used to encode most of the imaging reference information
  • attachment.url is DICOMweb URL used to convey:
    • Study UID
    • Series UID
    • SOP Instance UID
    • Frame list
  • Additional fields still needed for
    • Observation / Tracking UIDs
    • Segment ID
    • Image Region
  • Additional Document Type and Document Class codes needed to identify types of references
Σ1..*BackboneElement

Document referenced
Number of entries depends on type of reference

e.g.

  • Single image = single content item
  • Set of images = one content item per image
    • Or one content item per series
  • Set of frames in a single instance = single content item
  • etc.
Σ1..1AttachmentWhere to access the document
Σ0..1code

Mime type of the content, with charset etc.
MimeType (Required)

application/dicom

Σ0..1codeHuman language of the content (BCP-47)
Common Languages (Preferred but limited to AllLanguages)

0..1base64Binary

Data inline, base64ed

.. .. url

Σ1..1url

Uri where the data can be found

DICOMweb URL referencing either

  • Series
    • Solely applicable to use case where the set of images is an entire series
  • Instance
  • Frames
    • Can specify one or more frames in a multiframe instance

Note that URL is mandatory in this use case as it provides necessary context

Σ0..1integer64Number of bytes of content (if url provided)
Σ0..1base64BinaryHash of the data (sha-1, base64ed)
Σ0..1stringLabel to display in place of the data
Σ0..1dateTimeDate attachment was first created

      ... height

TU0..1positiveIntHeight of the image in pixels (photo/video)
TU0..1positiveIntWidth of the image in pixels (photo/video)
TU0..1positiveIntNumber of frames if > 1 (photo)
TU0..1decimalLength in seconds (audio / video)
TU0..1positiveIntNumber of printed pages
Σ0..1Coding

Format/content rules for the document
DocumentReference Format Code Set (Preferred)
Extension: Add one or more codes to cover DICOM images and SRs

.... observationUid

Σ0..*oidSelected observations in a DICOM SR

.... trackingUid

Σ0..*oidSelected tracking UIDs for longitudinal measurements in a DICOM SR

.... segmentID

Σ0..*positiveIntID of one or more segments in a DICOM Segmentation instance

.. .. imageRegion

Σ0..1BackboneElement

Specifies region in a DICOM image / frame

Only valid if it applies to all images / frames in the content item

..    .. regionType

Σ1..1Coding

Specifies the type of region

Need to make http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.10.5.html#sect_C.10.5.1.2 into a code table

  .   .. coordinateList[x]

Σ1..1
Specifies a sequence of coordinates.

 ..... 2dCoordinateList

Σ0..*decimalSpecifies a sequence of (x,y) coordinates

..      .. 3dCoordinateList

Σ0..*decimalSpecifies a sequence of (x,y,z) coordinates

Extension to context element

This was a previous attempt that added the additional attributes to the context element. This approach is superseded by the above work but is left on the page for reference.

Σ0..1BackboneElementClinical context of document

0..*Reference(Encounter)Context of the document content

0..*CodeableConceptMain clinical acts documented
v3 Code System ActCode (Example)
Σ0..1PeriodTime of service that is being documented

0..1CodeableConceptKind of facility where patient was seen
Facility Type Code Value Set (Example)

0..1CodeableConceptAdditional details about where the content was created (e.g. clinical specialty)
Practice Setting Code Value Set (Example)

0..1Reference(Patient)Patient demographics from source

0..*Reference(Any)Related identifiers or resources

0..*Reference(ServiceRequest | CarePlan)Procedure that caused this media to be created

... . imagingContext


0..1BackboneElementImaging context of document
Σ1..1oidStudy instance UID
Σ0..1Reference(ImagingStudy)Reference to ImagingStudy (could be excluded given the related attribute above)

         . .. seriesUid

Σ1..1oidSeries instance UID

.          .. instance

Σ0..*BackboneElementThe selected instances. If no instances are included the entire series is selected.
Σ0..1oidSOP class UID of instances

.      ... uid

Σ1..1Selected instance UIDs

.       .. frameNumber

Σ0..*integerSelected frames

     .  .. observationUid

Σ0..*oidSelected observations in a DICOM SR

     .  .. trackingUid

Σ0..*oidTracking UID for a measurement in a DICOM SR

.       .. imageRegion

Σ0..1BackboneElementSpecifies region in a DICOM image / frame

     .    .. regionType

Σ1..1Coding

Specifies the type of region

Need to make http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.10.5.html#sect_C.10.5.1.2 into a code table

         ... coordinateList[x]

Σ1..1
Specifies a sequence of coordinates.

             ... 2dCoordinateList

Σ0..*decimalSpecifies a sequence of (x,y) coordinates

.             .. 3dCoordinateList

Σ0..*decimalSpecifies a sequence of (x,y,z) coordinates
  • No labels

1 Comment

  1. Here are the extension and an example coded under context.  I will build out the entire Diagnostic report next and show the context per the PPT.ImagingContextExtension.StructureDefinition.jsonSampleDocRef.json