Pdf name UMLS_tutorial.

pdf "this doc is a copy and past from original one Tutorial T20 September 8, 2004

1 The Unified Medical Language System What is it and how to use it?
Olivier Olivier Bodenreider Bodenreider Jan Willis Jan Willis William Hole William Outline

1.1 What is UMLS?

How to use to use the the UMLS? Obtaining Obtaining a license license Remote Remote access access Knowledge Knowledge Source Server (UMLSKS) Source Server (UMLSKS) UMLSKS Application UMLSKS Application programming interface (API) interface (API) Local installation Local installation and customization (MetamorphoSys) Questions What is the UMLS? NOT END- USER Application

1.2 Introduction
Overview through an example The three UMLS Knowledge Sources UMLS Metathesaurus UMLS Semantic UMLS Lexicon Univifed medical language system Started in 1986 National Library of Medicine National Library of Medicine Long Long-term R&D project term R&D project Complementary to IAIMS

1.3 Overview through example

1.3.1 Addisons disease
Addison's disease is a rare endocrine disorder Addison's disease occurs when the adrenal glands do not produce enough of the cortisol For this reason, the disease is sometimes Called chronic adrenal insufficiency, or hypocortisolism

1.3.2 Adrenal insufficiency "Clinical variants"

Primary / Secondary Primary: lesion of adrenal glands themselves Secondary: inadequate Secondary: inadequate Acute / Chronic Isolated / Polyendocrine deficiency syndrome secretion of ACTH by the

1.3.3 Addisons disease Symptoms

Fatigue Weakness Low blood pressure Pigmentation of the skin (exposed and non Pigmentation of the ..

1.3.4 AD in medical vocabularies

Synonyms: different terms different Addisonian syndrome Bronzed disease Addison melanoderma Asthenia pigmentosa Primary adrenal deficiency Primary adrenal insufficiency Primary adrenocortical insufficiency Chronic adrenocortical insufficiency Contexts: different hierarchies eponym symptoms eponym

Clinical variants

1.4 Organize terms

Synonymous terms clustered into a concept Preferred term Unique identifier (CUI)

Organize concepts
One graph instead of instead of Multiple trees (Multiple inheritance)

B C

1.5 Relate to other concepts

Additional hierarchical relationships o link to other trees o make relationships explicit Non-hierarchical relationships Co-occurring concepts Mapping relationships

Relate to other concepts

1.6 Categorize concepts

High -level categories (semantic types) Assigned by the Metathesaurus editors Independently of the hierarchies in which these concepts are located

1.7 How do they do that?

Lexical knowledge Semantic pre-processing UMLS editors

1.7.1 Lexical knowledge

1.7.2 Semantic pre-processing

Metadata in the source vocabularies Tentative categorization Positive (or negative) evidence for tentative lexical features synonymy relations based on lexical features

1.7.3 UMLS editors: Additional knowledge

1.8 UMLS Summary

Synonymous Terms clustered into concepts Unique identifier Finer granularity Broader scope Additional hierarchical relationships Semantic categorization

10

2 UMLS Knowledge Sources

Metathesaurus o concepts o Inter-concept relationships Semantic Network o Semantic types o Semantic network relationships Lexical resources o SPECIALIST Lexicon SPECIALIST Lexicon o Lexical tools Lexical tools

1.9 Biomedical terminologies

1.9.1 General vocabularies
Anatomy drugs ( drugs ( RxNorm, First DataBank,Micromedex) , Micromedex) SPN) medical devices (UMD, SPN)

1.9.2 Several perspectives

clinical terms (SNOMED CT) Information sciences (MeSH, CRISP) administrative terminologies (ICD-9-CM, CPT CM, CPT-4) data exchange terminologies (HL7, LOINC)

1.9.3 Specialized vocabularies

nursing dentistry (CDT) oncology (PDQ) psychiatry (DSM, APA) adverse reactions (COSTART, WHO ART) primary care (ICPC)

1.9.4

Terminology of knowledge bases (AI/Rheum, DXplain, , QMR) The UMLS serves as a vehicle for the regulatory standards

11

1.10 Metathesaurus Basic organization

1.10.1Concepts
Synonymous terms are clustered into a concept Properties are attached to concepts, e.g., Unique identifier UniqueDefinition

1.10.2 Inter-concept relationships

Concepts are related to other concepts Properties are attached to relations, e.g., Properties are attached to relations, e.g.Type of relationship Type of relationship

1.10.3Addisons Disease: Addisons Disease: Concept

12

1.10.4Metathesaurus Concepts

13

1.10.5Cluster of synonymous terms of synonymous terms 1.10.6 1.10.7 1.10.8 1.10.9 1.10.10 1.10.11 1.10.12 1.10.13 1.10.14 1.10.15

1.11 Metathesaurus RELATIONSHIP

Symbolic relations: Symbolic relations: ~9 M pairs of concepts Statistical relations ~7 M pairs of concepts co-occurring concepts Mapping relations: Mapping relations: 100,000 pairs of concepts Categorization: Relationships between concepts: and semantic types from the Semantic Network

1.11.1

Symbolic relations

Relations:- Pair of atom identifiers Pair of atom identifiers Type Attribute (if any) List of sources (for type and attribute) Semantics of the relationship: Defined by its type [and attribute] Source transparency: The information is recorded at the atom level 14

1.11.1.1

Symbolic relations type

Hierarchal Parent / Child PAR /CH D RB/ RN

Broader / Narrower than Derived from hierarchies Siblings (children of parents)

SIB

Associative Other Various flavors of near Various flavors of near-synonymy Similar Source asserted synonymy Possible synonymy RL SY RO

RQ

1.11.1.2

Symbolic relations attributes

Hierarchical o isa -a ( is a kind-of) o part of Associative o location-of o caused by o treats

15

Cross -references (mapping)

16

17

1.11.1.3 1.12
1.12.1

Semantic Network
Semantic types

Tree structure 2 major hierarchies 2 major hierarchies Entity o Physical Object o Conceptual Entity Event o Activity o Phenomenon or Process

1.12.2Semantic network relationships

hierarchical (isa = is a kind of ) o among types Animal isa Organism Enzyme isa Biologically Active Substance o among relations treats isa affects non-hierarchical o Sign or Symptom Sign diagnoses Pathologic Function o Pharmacologic Substance treats Pathologic Function

Biologic Function hierarchy (isa)

18

19

Associative (non--isa) relationships

Semantic serve as high level categories Semantic assigned to Metathesaurus concepts, concepts, independently of their position in a hierarchy A relationship between 2 Semantic Types (ST) is a possible link between 2 concepts that have been assigned to STs The relationship may or may not hold at the concept level Other relationships may apply at the concept level Relationships can inherit semantics Relationships can inherit semantics

20

1.13
1.13.1

Lexical resources
SPECIALIST Lexicon
Content o English lexicon o Many words from the biomedical 200,000+ lexical items Word properties Word properties o morphology o orthography o syntax the lexical tools Used by the lexical tools

1.13.1.1

Morphology

Inflection o noun nucleus ,nuclei o verb cauterize, cauterizes, cauterized, cauterizing o adjective red, redder, reddest Derivation o verb noun cauterize --- cauterization o adjective noun red -- redness

1.13.1.2

Orthography

Spelling variants Spelling variants o oe/e oesophagus - esophagus o ae/e anaemia - anemia o ise/ize cauterise - cauterize o genitive mark Addison's disease Addison disease Addisons disease

1.13.1.3

Syntax

Complementation o verbs intransitive I'll treat. transitive He treated the patient. ditransitive He treated the patient with a drug. o nouns

prepositional phrase

Valve of coronary sinus 21

Position for adjectives

22

1.13.2

Lexical tools Lexical tools

Tomanage lexical variation in biomedical terminologies Major tools o Normalization o Indexes o Lexical Variant Generation program (lvg) Based on the SPECIALIST Lexicon engines Used by noun phrase extractors, search engines

1.13.2.1

Normalization
Hodgkins diseases, NOS Hodgkin diseases, NOS Hodgkin diseases hodgkin diseases, hodgkin diseases hodgkin disease disease hodgkin

e.g. and steps

Remove genitive Remove stop words Lowercase Strip punctuation Uninflect Sort words

Normalization Applications
Model for lexical resemblance Help find lexical variants for a term o Terms that normalize the same usually share the same LUI Help find candidates to synonymy among terms Help to map input terms to UMLS concepts

23

1.13.2.2

Indexes
Word index Word o word to Metathesaurus strings o one word index per language Normalized word index Normalized word index o normalized word to Metathesaurus strings o English only Normalized string index o normalized term to Metathesaurus strings o English only English only

1.13.2.3

Lexical Variant Generation program (lvg)

Tool for specialists (linguists) Performs atomic lexical transformations o generating inflectional variants o lowercase lowercase o Performs sequences of atomic transformations o a specialized sequence of transformations provides the normalized form of a term (the norm program)

24

How to use the UMLS?

Obtaining a license Remote access o Knowledge Knowledge Source Server (UMLSKS) o UMLSKS Application UMLSKS Application programming interface (API) Local installation and customization ( MetamorphoSys)

1.14 Obtaining a license

25

1.15 Remote access

UMLS Knowledge Source Server: http://umlsks.nlm.nih.gov Web search interface Application programming interface (API)

1.15.1UMLS Knowledge Source Server: Web search interface

26

27

1.16

28

1.17

29

30

31

32

33

34

1.17.1 1.17.2 Knowledge Source Server :Application Programming Interface

35

36

1.17.3

Local installation and ( MetamorphoSys)

Tool distributed with the UMLS Multi-platform Java software The UMLS installation and customization wizard o Installs Knowledge Sources to local o Subsets and customizes a local Metathesaurus

.i didnt complete implementation issues

37