ConText: An Algorithm for Identifying Contextual Features from Clinical Text

Date: October 22, 2007 Time: (All day)
Event Type: Lecture

Applications using automatically indexed clinical conditions must account for contextual features such as whether a condition is negated, historical or hypothetical, or experienced by someone other than the patient. We developed and evaluated an algorithm called ConText, an extension of the NegEx negation algorithm, which relies on trigger terms, pseudo-trigger terms, and termination terms for identifying the values of three contextual features. We will describe an evaluation of ConText's performance on Emergency Department reports and an evaluation of ConText's portability to five other report types.