Improved Patient Classification with Language Model Pretraining Over Clinical Notes

Clinical notes in electronic health records contain highly heterogeneous writing styles, including non-standard terminology or abbreviations. Using these notes in predictive modeling has traditionally required preprocessing (e.g. taking frequent terms or topic modeling) that removes much of the richness of the source data. We propose a pretrained hierarchical recurrent neural network model that parses minimally processed clinical notes in an intuitive fashion, and show that it improves performance for multiple classification tasks on the Medical Information Mart for Intensive Care III (MIMIC-III) dataset, improving top-5 recall to 89.7% (increase of 4.8%) for primary diagnosis classification and AUPRC to 35.2% (increase of 2.1%) for multilabel diagnosis classification compared to models that treat the notes as an unordered collection of terms, using no pretraining. We also apply an attribution technique to several examples to identify the words and the nearby context that the model uses to make its prediction, and show the importance of the words' context.