A system for automated lexical mapping

Abstract

Merging of clinical systems and medical databases, or aggregation of information from disparate databases, frequently requires a process where vocabularies are compared and similar concepts are mapped. Using a normalization phase followed by a novel alignment stage inspired by DNA sequence alignment methods, automated lexical mapping can map terms from various databases to standard vocabularies such as UMLS (Unified Medical Language System) and SNOMED (the Systematized Nomenclature of Medicine). This automated lexical mapping was evaluated using a real-world database of consultation letters from Children's Hospital Boston. The first phase involved extracting the reason for referral from the consultation letters. The reasons for referral were then mapped to SNOMED. The alignment algorithm was able to map 72% of equivalent concepts through lexical mapping alone. Lexical mapping can facilitate the integration of data from diverse sources and decrease the time and cost required for manual mapping and integration of clinical systems and medical databases.