Bulimia nervosa ( Eating Disorders) Mental Health Nursing.
I know just what you mean - Ontologies and their uses
1. I KNOW JUST WHAT YOU
MEAN - ONTOLOGIES
AND THEIR USES
Dave Parry
Dave.parry@aut.ac.nz
2. Agenda
What is an ontology ?
Why use them ?
How do you do that ?
Where next ?
3. Words
`When I use a word,' Humpty Dumpty said
in rather a scornful tone, `it means just
what I choose it to mean -- neither more
nor less.„
Lewis Carroll “Through the Looking
Glass – and what Alice found there”
Image from “Mathematics Enrichment”
4. Ontologies
“ Shared understanding of
a conceptualisation”
Deborah L. McGuinness
Controlled set of concepts
that are themselves linked
by concepts
5. What are they for ?
Mapping different terms to the same concept
Search and information retrieval – mapping terms
to concepts that are broader or narrower
Disambiguation – deciding WHICH concept you
mean
Finding and converting concepts – especially for
coding
6. Why use an Ontology ? example…
Pre-eclampsia
Gestational Proteinuric Hypertension
Toxaemia
GPH
PET
PE
All the same all map to.. pre-eclampsia 398254007 -
code
7. Why standardise ?
Synonyms and obsolescent terms
(HTLVIII vs HIV)
Needed for clinical activity
Communication
Data analysis
Coding
Reminders etc.
8. Why use it ?
In EHR systems, to replace/enhance free text.
To support reminders “If patient has allergy to
a drug code recorded then alert if drug is being
prescribed”
Converting free text to coded data….
9. What are they for ?
Mapping different terms to the same concept
Search and information retrieval – mapping
terms to concepts that are broader or narrower
Disambiguation – deciding WHICH concept
you mean
Finding and converting concepts – especially
for coding
10. Example
Medical ontology
A finger is “part of” a hand
A hand is “attached to” an arm
Soif your left arm is amputated – the fingers are not
present on that side of the body !
12. What are they for ?
Mapping different terms to the same concept
Search and information retrieval – mapping terms
to concepts that are broader or narrower
Disambiguation – deciding WHICH concept you
mean
Finding and converting concepts – especially for
coding
14. What are they for ?
Mapping different terms to the same concept
Search and information retrieval – mapping terms
to concepts that are broader or narrower
Disambiguation – deciding WHICH concept you
mean
Finding and converting concepts – especially for
coding
16. Systematized Nomenclature of
Medicine-Clinical Terms -
SNOMED CT
>600,000K Concepts
SNOMED CT is a clinical vocabulary currently
administered by the international health terminology
standards development organisation (IHTSDO)
http://www.ihtsdo.org/. Member countries are;
Australia, Canada, Denmark, Lithuania, The
Netherlands, New Zealand, Sweden, United Kingdom
and United States.
17. SNOMED CT
Very large clinical vocabulary that verges into being
an ontology
Includes Subsets to simplify disambiguation and
reduce search space
Mapping between SNOMED concepts and clinical
coding already exists.
18. General model
Extract SNOMED terms
Unstructured text EHR
Map to ICD 10 Send/receive messages
19. Ontology in Action
Original text Potential
fragments that
relate to
SNOMED terms
Potential
SNOMED
Concepts
Expanded
Selected Concepts SNOMED
Descriptions
20. Why is coding difficult ?
Experts don’t agree – even when a loose standard
of agreement is required (Chiang 2006)
SNOMED CT is very large and changes by 5-10%
each release
Data is used in ways that might be unfamiliar to
the originator
Ontologies can help resolve the meaning
Reliability of SNOMED-CT Coding by Three Physicians using Two Terminology Browsers
Michael F. Chiang, John C. Hwang, Alexander C. Yu, Daniel S. Casper, James J. Cimino, and Justin Starren
AMIA Annu Symp Proc. 2006; 2006: 131–135.
21. So what ?
Errors propagate through systems
SNOMED >ICD10 >DRG
Free text present in many places in systems.
Systems supporting coding may do better in
avoiding “Paper trail” errors (O’Malley 2005)
O'Malley, K. J., Cook, K. F., Price, M. D., Wildes, K. R., Hurdle, J. F., & Ashton, C. M.
(2005). Measuring diagnoses: ICD code accuracy.(International Classification of
Diseases). Health Services Research, 40(5), 1620(1620).
25. OWL – representing ontologies on
the web
Ontology web language
http://www.w3.org/TR/owl-ref/
Example :
http://owl.man.ac.uk/2003/why/latest/
26. Designing ontologies
Identify relationships between objects
Relationships depend on domain
Relationships can be simple “is-a” or extremely
complex e.g. “happens after all of the previous
events”
27. Take home message
Ontologies allow real-world
knowledge to be represented
in computer systems
Managing concepts is a key
task of health informatics
The semantic web allows us
to store knowledge on the
web – and even move to
wisdom on the web