Knowledge Extraction for Clinical Question Answering: Preliminary Results

The combination of recent developments in question answering research and the unparalleled resources developed specifically for automatic semantic processing of text in the medical domain provides a unique opportunity to explore complex question answering in the clinical domain. In this paper, we attempt to operationalize major aspects of evidence-based medicine in the form of knowledge extractors that serve as the fundamental building blocks of a clinical question answering system. Our evaluations demonstrate that domain-specific knowledge can be effectively leveraged to extract PICO frame elements from MEDLINE abstracts. Clinical information systems in support of physicians’ decisionmaking process have the potential to improve the quality of patient care in real-world settings.