UMLS Unified Medical Language System
Jan 03, 2016
UMLS
Unified Medical Language System
What is UMLS? A Unified knowledge representation
system Project of NLM
Large scale Distributed First launched in 1986
UMLS: the purpose To reduce the fundamental barriers
to the computer application in medicine: Lack of standard languages
• The disparity of terminologies used in different information resources and by different users
UMLS: approach Knowledge source for system
developers Metathesaurus The Semantic Network SPECIALIST lexicon and associated
lexical tools
UMLS Knowledge source
UMLS knowledge source: Core components:
Metathesaurs Semantic Network SPECIALIST lexicon and associated
lexical tools
Metathesaurus: Database of concepts Purpose:
link alternative names together identify useful relationships between different
concepts.
Metathesaurus: Organization
Basic principle: group alternative names together by
concept
Metathesaurs:Concept structure
Concept(CUI):
Term1 String 1 (SUI)(preferred)
(LUI) String 2 (SUI)
String 3 (SUI)
Term2 String 1
String 2
String 3
Metathesaurus:example TERM SOURCE TERM TYPE CODE
Zea mays MTH PN NOCODE Zea mays SNMI98 PT L-DB941 Zea mays CSP2000 ET 2340-8793 ZEA MAYS NDDF00 IN 006695 Corn <1> MTH MM U000077 Corn MSH2000 MH D003313 Corn SNMI98 SY L-DB941 Corn LCH90 PT U001161 corn AOD99 DE 0000013135 corn CSP2000 PT 2340-8793 Indian Corn MSH2000 EP D003313 Corn, Indian MSH2000 PM D003313 Maize MSH2000 EP D003313 Maize SNMI98 SY L-DB941 maize AOD99 NP 0000026523 maize CSP2000 ET 2340-8793
Metathesaurs relationships Hierarchical:
Broader(RB) Narrower(RN) Other related(RO) Parent (PAR) Child(CHD)
Non-hierarchical: LIKE(RL) Sibling(SIB) AQ QB
Metathesaurus: other data elements
Semantic types Additional Attributes
Definitions (source: MeSH, ..)
Metathesaurus:current status
In 2002 edition: Approximately 730,000 concepts 1.5 M concept names (terms) covers 60+ vocabularies and classifications,
such as:• MeSH(The medical subject heading)• DxPlain(MGH expert system)• UMDNS(Universal Medical Device Nomenclature
System)• PDQ2001(Physician Data Query Online System)• ICD2002(International Statistical Classification of
Disease)
UMLS Knowledge source: Core components:
Metathesaurs
Semantic Network Lexical tools
Semantic Network Provides categorization of all
concepts in Metathesaurus Currently including:
134 semantic types 54 semantic links (relationships)
Semantic Types (134): Sample Definition: tree structure Top notes:
“Entity” “Event”
Semantic relationships(54) hierarchical (isa = is a kind of)
Among types:• Plant isa organism• Activity isa event
Among relations:• Treats isa affects• Affects isa related to
non-hierarchical: Injuries or poisoning disrupts physiology functions Pharmacologic substance treats pathologic
functions
Hierarchical relations types:
Associative (non-isa) relatonships
Lexicon tools SPECIALIST lexicon Lexical tools
SPECIALIST Lexicon Content:
English lexicon 100,000+ Biomedical terms
Provide lexical information for lexical tools: Syntactic category Inflectional variations
UMLS Knowledge source: Core components:
Metathesaurs Semantic Network
SPECIALIST Lexicon and Lexical tools
Lexical tools To manage lexical variation in
medical terminologies Major tools
Normalization Indexed Lexical variation generation
Develop UMLS Applications
UMLS Distribution Knowledge source and associated lexical
tools Annual updates since 1990 Free, but requires a license agreement Full UMLS available:
CD-ROM ftp from the Knowledge Source Server Application Programming Interface(API) Knowledge Source Server
UMLS Application PubMed NLM Gateway NCBI Entrez Taxonomy
Names of all organisms that are represented in the genetic database with at least one nucleotide or protein substance.
Will be mapped to UMLS
Example: PubMedBronzed disease
UMLS tools
Addison’s disease(MesH term)
Search MeSH indexed database
Others: A UMLS-Based Knowledge Ac
quisition Tools for Rule-Based Clinical Decision Support System Construction
WizOrder