Supervised methods to extract clinical events from cardiology reports in Italian.

•We present supervised approaches for event extraction from Italian medical reports.•We explored recurrent neural networks (RNNs) that do not require elaborate features.•We created an annotated corpus to develop and evaluate the proposed approaches.•Using RNNs to extract clinical events from medical reports in Italian is promising.•We developed the first shared annotation guidelines for clinical text in Italian.•RNN classifiers can be useful for event extraction from non-English clinical text.