Unified Medical Language System UMLS NLM Presentation Theater

Unified Medical Language System® (UMLS®) NLM Presentation Theater MLA 2007 National Library of Medicine National Institutes of Health U. S. Dept. of Health and Human Services

The UMLS consists of Metathesaurus 1 million+ biomedical concepts from over 100 sources Semantic Network 135 broad categories and 54 relationships between them SPECIALIST Lexicon +Tools lexical information and programs for language processing 3 Knowledge Sources used separately or together

UMLS Objectives Began in 1986 as long-term R&D project § § Designed for systems developers Develop multi-purpose tools to enhance understanding of medical meaning across systems Overcome barriers to effective retrieval of machine-readable information § Overcome variety of ways the same concepts are expressed in machine readable and human language § Dr. Donald A. Lindberg Director National Library of Medicine

UMLS Uses § § § Information retrieval Thesaurus construction Natural language processing Automated indexing Electronic health records (EHR) Distribution mechanism for § HIPAA, CHI, PHIN regulatory standards § SNOMED CT

UMLS Releases § 3 -4 updates (releases) per year 3 Knowledge Sources in 3 sets of relational files § Tools: § § § Metamorpho. Sys – install and customize RRF Subset Browser Lexical tools including LVG Download from UMLSKS or on DVD

UMLS Access UMLS Knowledge Source Server (UMLSKS) § § § http: //umlsks. nlm. nih. gov/ Browse 3 Knowledge Sources Query with Java or XML-based APIs Download files and programs Access documentation and other resources Local installation § § § Use Metamorpho. Sys to install Knowledge Sources, customize Metathesaurus View customized subset with browser Use load scripts to load into database

UMLS License Agreements § Terms and conditions of use online § § § Semantic Network SPECIALIST Lexicon & Lexical Tools Metathesaurus § § Sign formal license agreement Additional restrictions apply to use of some sources as noted in the Appendix to the License Agreement

Metathesaurus 100+ general and specialized biomedical vocabularies § 17 languages (63% English) § 1 million+ concepts; 6 million+ names § 100 K+ relationships (hierarchical, semantic, statistical and mapping relationships) § § Distributed in a common electronic format

Metathesaurus Source Vocabularies Vary in purpose, structure and properties § Used in clinical, research, administrative, public health reporting § § All are sets of valid values § Thesauri, e. g. , Me. SH, CRISP, NCI § Statistical classifications, e. g. , ICD-9 -CM § Billing codes, e. g. , CPT, ABC Codes § Clinical coding systems, e. g. , SNOMED CT

SNOMED CT § Comprehensive clinical terminology § § Created by College of American Pathologists Ownership transferred to International Health Terminology Standards Development Organisation (IHTSDO) in April 2007 § § § 9 charter member countries includes U. S. NLM represents U. S. NLM distributes SNOMED CT to U. S. users in both native and UMLS formats

Metathesaurus Concepts Synonymous terms clustered into a concept § Unique identifier (CUI) is assigned § Source information preserved § Addison’s disease SNOMED CT Addison’s Disease Medline. Plus Addison Disease Me. SH Primary Adrenal Insufficiency Me. SH Primary hypoadreanlism Med. DRA syndrome, Addison … … C 0001403 PT PT PT EN LT 363732003 T 1233 D 000224 10036696 Addison’s disease

“The UMLS approach … assumes continuing diversity in the formats and vocabularies of different information sources and different information sources in the language employed by different elements of the biomedical community. It is not an attempt to build a single standard biomedical vocabulary. " Betsy L. Humphreys, Deputy Director, National Library of Medicine Humphreys, BL and PL Schuyler, The Unified Medical Language System: Moving beyond the vocabulary of bibliographic retrieval. In Broering NC, ed. High- Performance Medical Libraries: advanced information management for the virtual era. Westport (CT): Meckler; 1993, p. 33.

Semantic Network § 135 Semantic Types § Broad subject categories in 2 hierarchies § Assigned to all Metathesaurus concepts § 54 Semantic Relationships § Useful, important links between Types § Hierarchical “isa” and associative relations § Categorize the Metathesaurus § Enhance meaning of concepts

“Biologic Function” hierarchy (isa) Biologic Function 360 Physiologic Function 691 Organism Function 1528 Mental Process 1224 Organ or Tissue Function 2912 Cell Function 4417 Pathologic Function 9983 Molecular Function 13442 G e n e ti c Function 1340 Cell or Molecular Dysfunction 1276 Disease or Syndrome 67716 Mental or Behavioral Dysfunction 5691 Experimental Model of Disease 72 Neoplastic Process 19436

Semantic Relations: between types u Disease or Syndrome associated_with Finding u Disease or Syndrome result_of Pathologic Function u Body Part, Organ, or Organ Component location_of Disease or Syndrome u Hormone affects Disease or Syndrome Hormone causes Disease or Syndrome Hormone complicates Disease or Syndrome

SPECIALIST Lexicon and Lexical Tools § English lexicon of 300 K+ common words and biomedical terms § Lexical records encode information on: § § § Syntax Morphology Orthography § Used with associated lexical tools § in Metathesaurus production § in natural language processing applications

SPECIALIST Lexicon Lexical Entry {base=disease entry=E 0023270 cat=noun variants=reg variants=uncount compl=pphr(of, np|bone|) compl=pphr(of, np|breast|) compl=pphr(of, np|liver|) compl=pphr(of, np|ovary|)} Base form Unique identifier Part of speech Lexical variants Prepositional phrase complements

Lexical Tools § Manage lexical variation in biomedical terminologies and text § Used separately or with SPECIALIST Lexicon § Perform transformations selected and ordered by users § 3 primary programs: normalizer, word index generator, lexical variant generator http: //umlslex. nlm. nih. gov/lvg/current/

Normalization 1 Hodgkin’s diseases, NOS Remove genitive Remove stop words Lowercase Strip punctuation Uninflect Hodgkin diseases, NOS Hodgkin diseases, hodgkin diseases hodgkin disease Sort words disease hodgkin

Normalization 2 Hodgkin Disease HODGKINS DISEASE Hodgkin's Disease, Hodgkin's, disease HODGKIN'S DISEASE Hodgkin's disease Hodgkins Disease Hodgkin's disease NOS Hodgkin's disease, NOS Disease, Hodgkins Diseases, Hodgkins Diseases Hodgkins disease hodgkin's disease Disease, Hodgkin normalize disease hodgkin

UMLS Knowledge Source Server (UMLSKS) Home Page From top links or buttons § Search 3 Knowledge Sources From sidebar § Downloads § Documentation § Resources

UMLS Documentation and Support § UMLS Home Page § § UMLSKS § § http: //umlsks. nlm. nih. gov NLP and Lexical Tools § § http: //umlsinfo. nlm. nih. gov/ http: //lexsrv 3. nlm. nih. gov/SPECIALIST/index. h tml NLM customer service § Email: custserv@nlm. nih. gov

Summary § 3 Knowledge Sources § § § Metathesaurus Semantic Network Lexicon + Lexical Tools § Metamorpho. Sys § UMLSKS install, customize, browse, query, download § License Agreement

Thank you