Automatic identification of negated concepts in narrative clinical reports

Substantial medical data such as discharge summaries and operative reports are stored in textual form. Databases containing free-text clinical narratives reports often need to be retrieved to find relevant information for clinical and research purposes. Terms that appear in these documents tend to appear in different contexts. The context of negation, a negative finding, is of special importance, since many of the most frequently described findings are those denied by the patient or subsequently "ruled out." Hence, when searching free-text narratives for patients with a certain medical condition, if negation is not taken into account, many of the documents retrieved will be irrelevant. In this paper we examine the applicability of machine learning methods for automatic identification of negative context patterns in clinical narratives reports. We suggest two new simple algorithms and compare their performance with standard machine learning techniques such as neural networks and decision trees. The proposed algorithms significantly improve the performance of information retrieval done on medical narratives.