DICOM and Image Ontology Image Ontology Workshop 2006 Stanford University

1
DICOM and Image Ontology Image Ontology
Workshop 2006Stanford University
B I O M E D I C A L O N T O L O G Y

• Anand Kumar MD, PhD, MBA
• IFOMIS, Univ. of Saarland, Germany.
• Siemens Medical Solutions, Germany.

2
B I O M E D I C A L O N T O L O G Y
DICOM
Digital Imaging and Communications in Medicine
Joint standard from American College of Radiology
and National Electronic Manufacturers
Association
A standard method for transferring images and
associated Information between devices
manufactured by vendors
Addresses the needs particularly of Radiologists
3
B I O M E D I C A L O N T O L O G Y
Basic definitions
Information Object (Class) An abstraction of a
real information entity (e.g., CT Image,
Structured Report, etc.) which is acted upon by
one or more DICOM Commands Synonym of
Information Object Definition (IOD)
Service Class A structured description of a
service which is supported by cooperating DICOM
Applications using specific DICOM Commands
acting on a specific class of Information Object.
4
(No Transcript)
5
B I O M E D I C A L O N T O L O G Y
Information Object Classes
Normalized Information Object Class Includes
only those Attributes inherent in the real-world
entity represented.
Composite Information Object Class May
additionally include Attributes which are
related to but not inherent in the real-world
entity. E.g. Patient Name together with Computed
Tomography Image Information Object Class
Expresses the communication requirements of
images where image data and related data needs
to be closely associated.
6
B I O M E D I C A L O N T O L O G Y
Entity and Relationship
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
B I O M E D I C A L O N T O L O G Y
DICOM Relations and Attributes (Extra)
General Study Module Has-study-date Has-study-tim
e Has-referring-physician Has-accession-number Has
-physicians-of-record Has-physicians-reading-study
Has-reference-study Has-procedure-code
17
B I O M E D I C A L O N T O L O G Y
DICOM Relations and Attributes (Extra)
General Series Module Has-modality Has-laterality
Has-series-date Has-series-time Has-performing-ph
ysician Has-protocol Has-operator Has-reference-pe
rformed-step Has-related-series Has-body-part-exam
ined Has-patient-position Has-smallest-pixel-value
-in-series Has-Largest-pixel-value-in-series Has-I
maging-Service-Request Has-performed-procedure-ste
p Has-performed-procedure-step-start-time Has-perf
ormed-procedure-step-start-date Has-performed-prot
ocol-code
18
B I O M E D I C A L O N T O L O G Y
DICOM Relations and Attributes (Extra)
General Image Module Has-patient-orientation Has-
content-date Has-content-time Has-image-type Has-A
cquisition-date Has-Acquisition-time Has-reference
d-image-sequence Has-derivation Has-source-image H
as-referenced-waveform Has-images-in-acquisition
(could be put in the acquisition process
module) Has-quality-control-image Has-burned-in-an
notation Has-lossy-image-compression Has-lossy-ima
ge-compression-ratio
19
B I O M E D I C A L O N T O L O G Y
DICOM Relations and Attributes (Extra)
Image Pixel Module Has-samples-per-pixel Has-photo
metric-interpretation Has-rows Has-columns Has-bit
s-allocated Has-bits-stored Has-high-bit Has-pixel
-representation Has-pixel-data Has-planar-configur
ation Has-pixel-aspect-ratio Has-smallest-image-pi
xel-value Has-largest-image-pixel-value Has-red-pa
lette-color-lookup-table-descriptor Has-blue-palet
te-color-lookup-table-descriptor Has-green-palette
-color-lookup-table-descriptor Has-red-palette-col
or-lookup-table-data Has-blue-palette-color-lookup
-table-data Has-green-palette-color-lookup-table-d
ata
20
B I O M E D I C A L O N T O L O G Y
Multiframe Images
Mutiframe Images Large number of images can be
sent as one object All attributes equal to all
images or their groups sent once Modality
independent functional groups Modality dependent
functional groups
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
B I O M E D I C A L O N T O L O G Y
What should Image Ontology Cover?
Image Series Study Patient Visit Secondary
Capture Image Based Reporting Image Based
Diagnosis Radiotherapy Clinical Trials
27
B I O M E D I C A L O N T O L O G Y
DICOM Structured Reporting
Template Specifications which include Concept
Name, Requirement, Value Type, Value
Multiplicity, Value Set Restriction, Relationship
Types
28
B I O M E D I C A L O N T O L O G Y
29
B I O M E D I C A L O N T O L O G Y
Examples
Reporting
DICOM Image-Related Object
30
B I O M E D I C A L O N T O L O G Y
Advice
Cover the depth to the extent DICOM does Use
DICOM types Reclassify them Provide correct
relations
FMA is a reference ontology too FMA is not small
31
DICOM and Image Ontology Image Ontology
Workshop 2006Stanford University
B I O M E D I C A L O N T O L O G Y

• Anand Kumar MD, PhD, MBA
• IFOMIS, Univ. of Saarland, Germany.
• Siemens Medical Solutions, Germany.