Unified Medical Language System® (UMLS) is a coding system that facilitates communication between various medical vocabularies by integrating terminology, classification, and standard codes. Furthermore, the UMLS package includes associated software that aids the creation of more interoperable software and systems.
In other words, UMLS is used for the development of computer systems that need to “understand” medical terminology across various specialty fields.
It is proficient in information retrieval and processing of common medical language and the creation of patient and research data. For this purpose, it has been widely used by PubMed, MedlinePlus, Connect, and ClinicalTrials.org, among others.
UMLS has a broad range of use, and it facilitates the linking of codes between doctors, pharmacies, clinicians, healthcare managers, and insurance companies.
In clinical trials, UMLS facilitates mapping between technologies and the development of terminology for specific research.
UMLS is produced and distributed by the National Library of Medicine (NLM) as a composite system that contains three UMLS Knowledge Sources and associated software tools.
UMLS Knowledge Sources
UMLS Knowledge Sources are not optimized for a particular organization or application. Still, they are multi-purposed and as such, they can perform various functions and integrate different types of information.
They are applied in various healthcare-related fields, such as reviewing the literature, keeping healthcare records, and organizing public health data.
There are three UMLS Knowledge Sources: the Metathesaurus, the Semantic Network, and the SPECIALIST Lexicon.
It is a multilingual vocabulary database that is organized by meaning. It communicates with other source vocabularies and identifies relationships between terms and concepts.
The Metathesaurus is built from many electronic data sets, and it contains structures, such as patient care, health service billing, indexing of biomedical literature, clinical searches, and public health statistics.
Metathesaurus communicates terms with over 150 medical dictionaries, such as ICD-9-CM, ICD-10-CM, SNOMED CT, CPT, MeSH, RxNorm, and LOINC.
All concepts within the Metathesaurus are assigned to at least one semantic type from the Semantic Network, which disambiguates and categorizes these concepts for informational systems.
Metathesaurus vocabularies are available in one of two major UMLS database formats:
- Rich Release Format (RRF), or
- Original Release Format (ORF).
Metathesaurus interprets terms and codes from other vocabularies, including hierarchies, definitions, and other relationships and attributes.
The Semantic Network
The Semantic Network consists of:
- Semantic categories, which classify the terms in the Metathesaurus. Major semantic groups refer to organisms, anatomical structures, biologic function, chemicals, events, and physical objects.
- Relationships that exist between these categories and describe their functions.
The SPECIALIST Lexicon
The SPECIALIST Lexicon provides the word usage information needed for the SPECIALIST Natural Language Processing (NLP) System. It describes every word used in biomedical terminology.
Each word or term in this knowledge source contains the syntactic, morphological, and orthographic information needed by the SPECIALIST NLP System.
In other words, the SPECIALIST Lexicon system is a sizable syntactic database comprised out of biomedical and general English, with a set of tools for normalizing strings, variants, and indexes.
UMLS Software Tools
Software tools that accompany UMLS are used by developers to build and customize the use of UMLS for particular purposes.
There are two main tools: Lexical Tools and MetamorphoSys.
MetamorphoSys is the UMLS installation wizard and customization tool for Metathesaurus. It also installs all of the UMLS Knowledge Sources and creates optional vocabulary subsets.
UMLS Terminology Services (UTS)
It is a web interface for the UMLS Knowledge Sources.