Combining semantic and lexical methods for mapping MedDRA to VCM icons

Jean-Baptiste Lamy (1) , Rosy Tsopra (2, 3)

1 LIMICS - Laboratoire d'Informatique Médicale et Ingénierie des Connaissances en e-Santé
2 HEGP - Hôpital Européen Georges Pompidou [APHP]
3 CRC (UMR_S_1138 / U1138) - Centre de Recherche des Cordeliers

VCM (Visualization of Concept in Medicine) is an iconic language that represents medical concepts, such as disorders, by icons. VCM has a formal semantics described by an ontology. The icons can be used in medical software for providing a visual summary or enriching texts. However, the use of VCM icons in user interfaces requires to map standard medical terminologies to VCM. Here, we present a method combining semantic and lexical approaches for mapping MedDRA to VCM. The method takes advantage of the hierarchical relations in MedDRA. It also analyzes the groups of lemmas in the term's labels, and relies on a manual mapping of these groups to the concepts in the VCM ontology. We evaluate the method on 50 terms. Finally, we discuss the method and suggest perspectives.

Mots clés: Terminology as Topic; MedDRA; Computer Graphics; Nonverbal Communication; Icons

Domaines: Santé publique et épidémiologie; Intelligence artificielle [cs.AI]

Jean-Baptiste Lamy, Rosy Tsopra. Combining semantic and lexical methods for mapping MedDRA to VCM icons. Studies in Health Technology and Informatics, 2018, 247. ⟨hal-01803759⟩