Building a specialized lexicon for breast cancer clinical trial subject eligibility analysis

A natural language processing (NLP) application requires sophisticated lexical resources to support its goals. Different solutions, such as dictionary lookup and MetaMap, have been proposed in the healthcare informatics literature identify disease terms with more than one word (multi-gram named entities). Although a lot of work has done identification protein- gene-named entities biomedical field, not much research on recognition resolution terminologies clinical trial subject eligibility analysis. In this study, we develop specialized lexicon for improving NLP text mining analysis breast cancer domain, evaluate it by comparing Systematized Nomenclature Medicine Clinical Terms (SNOMED CT). We use hybrid methodology, which combines knowledge domain experts, from multiple online dictionaries, sample trials. Use our methodology introduces 4243 unique items, increase bigram entity match 38.6% trigram 41%. Our lexicon, adds significant number new terms, is very useful matching patients trials automatically based matching. Beyond matching, developed study could serve foundation future applications.