Extracting medication information from French clinical texts.

摘要: Much more Natural Language Processing (NLP) work has been performed on the English language than any other. This general observation is also true of medical NLP, although clinical processing needs are as strong in other languages they English. In specific subdomains, such drug prescription, expression information can be closely related across different languages, which should help transfer systems from to languages. We report here implementation a medication extraction system extracts drugs and French texts, basis an approach initially designed for within framework i2b2 2009 challenge. The relies specialized lexicons set rules. A first evaluation 50 annotated texts obtains 86.7% F-measure, level higher original close work. shows that same rule-based applied with similar performance. further discuss directions improving both systems.