Towards an OWL-based framework for extracting information from clinical texts

Nate Blaylock, William de Beaumont, James F. Allen, Hyuckchul Jung

Abstract: In this paper, we present our ongoing work towards an OWL-based framework for extracting a variety of information (including patient history) from clinical texts. Our framework integrates a well-known natural language processing (NLP) system by converting its ontology and output logical form interpretation into the Web Ontology Language (OWL). The OWL-based Semantic Query-Enhanced Web Rule Language (SQWRL) is then used as a platform for authoring Semantic Web-aware rules for extracting information of interest from the OWL knowledge based created from parsing a clinical report. We also describe our ongoing work on using this system for extracting a timeline-based patient medical record from the history of present illness section of clinical texts.