Identifying Patients with Pneumonia from Free-Text Intesive Care Unit Reports

Clinical research studying critical illness phenotypes relies on the identification of clinical syndromes defined by consensus definitions. Pneumonia is a prime example. Historically, identifying pneumonia has required manual chart review, which is a time and resource intensive process. The overall research goal of our work is to develop automated approaches that accurately identify critical illness phenotypes. In this paper, we describe our approach to the identification of pneumonia from electronic medical records, present our preliminary results, and describe future steps.