Table 5.

Summary of inconveniences and advantages of the different methods.

Algorithms	What is already available?	Characteristics of the algorithm
Using MedDRA-to-SNOMED CT mappings from UMLS Metathesaurus	MedDRA-to-SNOMED CT mappings in UMLS Metathesaurus	Available mappings are used to retrieve SNOMED CT concepts associated to a MedDRA term. Properties are added to the formal definition according to the SNOMED CT concept position in the hierarchy. It illustrates Scenario 6. “Reusing, Merging and Re-engineering Ontological Resources” of the NeON methodology. The algorithm is automatic, except when several SNOMED CT concepts are available which requires expert selection.
Using other MedDRA-to-SNOMED CT mapping resources	Nadkarni and Darer’s propositions of mappings	This is an expert-based process entirely manual of validation and refinement that illustrates Scenario 2. “Reusing and Re-engineering Non-Ontological Resources” of the NeON methodology but benefits from a non-ontological resource that expedites formal definition of MedDRA terms compared with manual definitions.
Using a syntactic decomposition algorithm on complex MedDRA terms	–	This algorithm is automated and developed ad hoc. It illustrates one of the ontology support activities, “knowledge acquisition,” and exploits hidden semantics as proposed by Third (2012). It is limited to complex MedDRA terms.
Automatic lexical enrichment methods	–	This algorithm is based on substring search. It necessitates defining beforehand substrings that may be associated to a SNOMED CT concept. Review of the algorithm’s proposal is mandatory in order to check that substrings allow associating with relevant SNOMED CT concepts. It also illustrates Ontology support activity “knowledge acquisition” and exploits hidden semantics.
Manual definition	–	This process is manual and expert based, but the Ci4SeR tool suggests definitions on the basis of definitions already available for siblings and parents of a MedDRA PT. Such approach addresses both “knowledge acquisition” and “ontology validation” within the Ontology support activities.