Source Image Sequence Attribute
| Tag | (0008,2112) |
|---|---|
| Type | Required (1) |
| Keyword | SourceImageSequence |
| Value Multiplicity | 1 |
| Value Representation | Sequence (SQ) |
A Sequence that identifies the Images that were used to derive this Image.
One or more Items shall be included in this Sequence.
See Section C.12.4.1.2 and Section C.8.17.14.1.1 for further explanation.
Section C.12.4.1.2
C.12.4.1.2 Source Image Sequence
If an Image is identified to be a Derived Image (see Section C.7.6.1.1.2 “Image Type”), Source Image Sequence (0008,2112) is an optional list of source images used to create the Derived image. Source Instance Sequence (0042,0013) is an optional list of non-image source Instances that were used to create this Instance. Source Image Sequence (0008,2112) and/or Source Instance Sequence (0042,0013) may be used whether or not there is a description of the way the Instance was derived in Derivation Description (0008,2111) or Derivation Code Sequence (0008,9215).
Images shall not be referenced by Source Instance Sequence (0042,0013).
Note
Multiple Items may be present within Source Image Sequence (0008,2112) and/or Source Instance Sequence (0042,0013), in which case either:
-
those Instance were combined to make the derived Instance (e.g., multiple source images to make an MPR or MIP), or
-
each of the Items represents a step in the successive derivation of an Instance (e.g., when an image has had successive lossy compression steps applied to it),
-
some combination of the above.
The Purpose of Reference Code Sequence (0040,A170) and the Attributes within the referenced Instances themselves may be used to determine the history of the derivation, which is not otherwise explicitly specified.
Section C.8.17.14.1.1
C.8.17.14.1.1 Source Image Sequence
An OCT en face image is derived from images obtained using OCT technology. The Source Image Sequence (0008,2112) shall convey the SOP Instances used to derive this en face SOP Instance.
If Attribute Purpose of Reference Code Sequence (0040,A170) is set to (128250, DCM, "Structural image for image processing"), the Source Image Sequence will reference an Ophthalmic Tomography SOP Instance.
If Attribute Purpose of Reference Code Sequence (0040,A170) is set to (128251, DCM, "Flow image for image processing"), the Source Image Sequence will reference an Ophthalmic Optical Coherence Tomography B-scan Volume Analysis SOP Instance.
A typical example of the image processing stages performed to generate en face images is shown in Figure C.8.17.14-1.
Figure C.8.17.14-1. Example of the Image Process Performed to Generate En Face Images
Figure Legend:
-
OCT proprietary B-scan data (possibly a DICOM Raw Data Instance)
-
Volumetric structural ophthalmic tomography image (Ophthalmic Tomography Image Instance)
-
OCT angiographic flow volume information (Ophthalmic Optical Coherence Tomography B-scan Volume Analysis Instance)
-
OCT surface mesh (Surface Segmentation Instance)
-
Structural en face image (Ophthalmic Optical Coherence Tomography En Face Image Instance)
-
En face angiographic flow image (Ophthalmic Optical Coherence Tomography En Face Image Instance)
Stage 1: OCT technology is used to acquire a volumetric data set from a retinal region of interest. This volumetric data set (A) consists of multiple B-scans in a raster pattern, and multiple frames are acquired at each B-scan location. The B-scans are acquired in the manufacturer's proprietary format for analysis and storage. If this information is stored in DICOM, it can use the Raw Data Storage SOP Class.
Stage 2: The OCT proprietary B-scan data (A) (or DICOM Raw Data SOP Instance) is then analyzed to derive the volumetric structural ophthalmic tomography image (B). From (B) one or more OCT surface meshes (D) are generated to delineate the anatomical boundaries. The difference in signal between the frames of each individual B-scan is analyzed to produce the OCT angiographic flow volume information (C).
Stage 3: Clinicians typically make their assessment based upon two types of OCT en face images. The structural OCT en face image (E) is derived by using pixel information in (B) and two surface meshes (D). The OCT angiographic flow en face image (F) may be derived using the OCT angiographic flow volume information (C) and the OCT surface mesh (D).
En face images are typically derived by the acquisition modality that generated the Ophthalmic Tomography Image, Surface Segmentation and Ophthalmic Optical Coherence Tomography B-scan Volume Analysis SOP Instances or by image workstations that received the respective Ophthalmic Image, Surface Segmentation and Ophthalmic Optical Tomography B-scan Volume Analysis SOP Instances via DICOM Storage.
Note
Image workstations receiving ophthalmic tomography images may choose to evaluate the structural ophthalmic tomography image and generate a different set of segmented surfaces than defined by an acquisition device. The surface segmentation information can be stored in a separate Surface Segmentation SOP Instance.
The Ophthalmic Tomography Image, Surface Segmentation, Ophthalmic Optical Coherence Tomography B-scan Volume Analysis and the Ophthalmic Optical Coherence Tomography En Face Image SOP Instances all reside in different DICOM Series. They share the same spatial Frame of Reference which is identified in Attribute Frame of Reference UID (0020,0052) (i.e., the value of Frame of Reference UID (0020,0052) is the same in each SOP Instance). Figure C.8.17.14-2 illustrates the relationships between the OCT angiography based SOP Instances.
Figure C.8.17.14-2. Relationships Between OCT-A Based SOP Instances