HEALTH SCIENCES TERMINOLOGY

Broad Issues:

• each domain has its own unique vocabulary
• health sciences education, research and clinical care is by nature interdisciplinary
• communication between the individuals and disciplines is at the heart of the practice of healthcare
• provision of integrated information resources requires some "translation" to happen

Vocabulary problems and opportunities:

• synonyms  (aspirin, Bayer)
• narrower/broader concepts (mammal, rodent, mouse, ....)
• degrees of specificity (cell, nucleus, nucleoli...)
• difference between "representation" of a concept and "describing" a thing (e.g. SNOMED versus Medline)
• computer "understanding" of terminology -- e.g. mapping so that automated reminder systems can generate action (e.g. launch a specific search of Medline) based on a specific diagnosis in the medical record

Vocabulary: a finite enumerated set of terms intended to convey information unambiguously

unfortunately a single, integrated, comprehensive health sciences vocabulary does not exist!

Term (definition): symbolic representation of a single concept (and optionally, some of the concept's attributes)

Goal of retrieval:

• unambiguous retrieval
• complete = ALL the data I want
• specific = ONLY the data I want
• reliable = same data for every similar query

Existing vocabulary examples:

• ICD-9/10 - International Classification of Diseases
World Health Org. for collecting health stats. -- orig. for epidemiology
• SNOMED -- Systematized Nomenclature of Medicine
• College of American Pathologists wrote because ICD-9 did not deal well enough for pathology
• MeSH - Medical Subject Headings
  (accessible from PubMed homepage)
• National Library of Medicine developed to described the medical literature

Unified Medical Language System -- A Potential Solution?

• National Library of Medicine Initiative
• Goal: to facilitate the retreival and integration of information from many machine-readable information sources, including descriptions of the biomedical literature, patient records, factual databanks, medical knowledge bases etc.
• It is NOT an attempt to impose either a single standard vocabulary, a single standard record format or a single medical knowledge base on the biomedical community

UMLS Knowledge Sources Components:

• Metathesaurus:  Annual editions of the Metathesaurus have been distributed since 1990. NLM moved to quarterly updates in 2002. The November 2003AC edition includes 975,354 concepts and 2.4 million concept names in over 100 biomedical source vocabularies, some in multiple languages.

   

  • links terminology and concepts from a range of vocabularies and classifications
  • transcends individual thesauri by virtue of links
  • original meaning of term from source vocabulary is always preserved
  • tool for transforming ideas into words

   

• Semantic Network (began 1990)
• provides consistent view of the concepts represented in the metathesaurus
• a structure for categorizing objects in the biomedical domain (e.g. "disease or syndrome" is a "pathologic function" is a "biologic function" is a "natural phenomenon or process"
• attempts to impart "common sense" knowledge to computers -- allowing them to reason and draw conclusions
• Information Sources map (began 1991)
• describes computerized biomedical  information sources
• provides users with a path to the most appropriate databases based on the particular query posed.
• Specialist Lexicon (began 1994) -

• Lexical information: for each word or term records the syntactic, morphological, and orthographic information needed by the SPECIALIST natural language processing system.
•  Coverage includes both commonly occurring English words and biomedical vocabulary

• Recommended further reading:

  Lindberg, D.A.B., Humphreys B.L., McCray A.T. The Unified Medical Language System. Meth. Inform. Med 32:281-291, 1993.

  UMLS Factsheet

   

 REV:  1/12/06