Detecting concept relations in clinical text: Insights from a state-of-the-art model

[Display omitted] ► Discuss a top-ranked model for detecting concept relations in real clinical text. ► Knowledge sources of various nature play different roles but cooperated very well. ► Unrecognized concepts dramatically impair performance, raising more research concerns. ► Labeled data are still insufficient, while leveraging unlabeled data helps. This paper addresses an information-extraction problem that aims to identify semantic relations among medical concepts (problems, tests, and treatments) in clinical text. The objectives of the paper are twofold. First, we extend an earlier one-page description (appearing as a part of [5]) of a top-ranked model in the 2010 I2B2 NLP Challenge to a necessary level of details, with the belief that feature design is the most crucial factor to the success of our system and hence deserves a more detailed discussion. We present a precise quantification of the contributions of a wide variety of knowledge sources. In addition, we show the end-to-end results obtained on the noisy output of a top-ranked concept detector, which could help construct a more complete view of the state of the art in the real-world scenario. As the second major objective, we reformulate our models into a composite-kernel framework and present the best result, according to our knowledge, on the same dataset.