Table 3.

Basic terms and definitions.

Term	Definition	Note
representational unit (RU)	A language-independent denotator that corresponds to a node in a terminology or ontology.	RUs are typically identified by a code and a human-readable label. Optionally, their meaning is described by scope notes and by textual and logic-based elucidations or definitions.
term	A linguistic expression (ranging from a single word to a phrase) that belongs to a domain-specific vocabulary in a given natural language.	We also distinguish labels from terms (see “label”).
label	An often artificially constructed term that aims to attach a maximum of unambiguous meaning to an RU.	For example, “Biopsy of head and neck structure” is an unambiguous label but might not be a term that would be commonly used by practitioners.
concept	No definition given because “concept” is used in numerous, partly contradictory senses [17]. Despite the immense popularity of this word, we recommend always using it with an attribute, e.g., “SNOMED CT concept”, in order to avoid ambiguous interpretations (in which case it is referencing an RU in the associated artifact).	According to the community, it is used in the sense of “entity of thought” (encompassing classes, binary relations, and individuals in SNOMED CT; classes and individuals in UMLS), “information template” (clinical model community), “unary predicate” (logic), “universal” (ontology).
class	A group of things that share some properties.	Linguistic expressions (terms, labels, scope notes, definitions), as well as formal axioms, describe the class and criteria for membership. Most RUs (nodes) in ontologies denote classes. Synonym: universal
individual	A single thing in a domain.	Individuals are members of classes.Synonym: particular.
terminology	An information artifact that includes a “set of designations belonging within a discipline”.	Normally, these designations are “terms”, i.e., units of human language, but codes might also be encompassed. Given the broad use of “terminology” in biomedical informatics as well as in biomedical science (beyond the common use of “terminology” in the terminology world), we recommend always using it together with qualifying adjectives.
reference terminology	A terminology that organizes RUs in a domain, with human readable, maximally self-explaining labels and potentially formal or textual definitions or scope notes.	Reference terminologies are uncommitted to any specific purpose. Their representational units are often named “concepts”, e.g., SNOMED CT concepts.
aggregation terminology	A terminology where RUs are systematically organized in single nonoverlapping hierarchies, enhanced by classification rules.	Aggregation terminologies are also known as classifications, e.g., the WHO classifications such as the International Classification of Diseases. Aggregation terminologies are meant for specific purposes like data aggregation and ordering. Synonym: classification.
user interface terminology	A terminology containing terms used in written and oral communication within specific contexts determined by language, dialect, application, and user groups.	User interface terminologies either lack semantic import altogether or acquire their necessary semantic import by linkage to reference terminologies/ontologies or aggregation terminologies. They are often ambiguous and, therefore, not just alternative labels. User interface terminologies obey less strict organizing principles; terms are grouped topically rather than ontologically.
thesaurus	An informal terminology that groups together words and terms according to similarity of meaning.	In biomedical informatics, there is a tendency to extend the meaning of the word "ontology" also to thesauri (e.g., MeSH, NCIt, UMLS), which we recommend to avoid.
formal ontology	A representational artifact, comprising a taxonomy as proper part, whose representational units are intended to designate some combination of universals, defined classes, and relations between them.	The use of the word “ontology” should be limited to carefully engineered, principled, and computable models of meaning. As such, ontologies can be described as formally founded reference terminologies. Thesauri (e.g., MeSH) or simple data models (i2b2) are not ontologies in our sense. We avoid the use of “ontology” in the sense of the Obrst semantic spectrum [18]. To enhance clarity, we recommend the use of this composed term “formal ontology”.
standard	An information artifact developed in community-driven consensus processes that specifies uniform criteria, methods, processes, and/or practices for a certain domain.	The term “standard” is also used in a broader sense for specifications that adopt a “de facto” standard status due to acceptance by a large public or market forces.The term “standard” may be applied (i) to single artifacts, which may be huge (e.g., SNOMED CT) or tiny (a single ISO13606 or openEHR archetype), (ii) processes for creating artifacts that connect into a larger whole, and (iii) shells that support the creation of (i) by (ii) by numerous distributed parties.
quasi-standard	A specification that is not the outcome of a standardization process but is nevertheless accepted by a larger community.	Due to their large acceptance, quasi-standards can be relatively safely referred to where real standards do not exist.Synonym: de facto standard
information model	An information artifact for a specific domain like healthcare, by which a bounded set of facts, assertions, and instructions are expressed to meet a specified requirement, typically that of an implementation.	Elements of an information model can be instantiated to create persistent and/or in-memory data.
reference model	A model that typically defines a logical model of data based on very generic entities (classes) and data types.	In the healthcare domain, reference models are often published as standards [19,20].
semantic interoperability	The ability of the flow of information between different components within the same or different systems or institutions to be meaning-preserving.	Semantic interoperability attracts growing attention in the area of biomedical semantics.
scale	The level of granularity in describing physical entities.	Macroscale: the patient—microscale: genes, proteins, etc.