Computerized Text Analysis to Enhance Automated Pneumonia Detection

Sylvain DeLisle, Tariq Siddiqui, Adi Gundlapalli, Matthew Samore, Leonard D'Avolio


Information about disease severity could help with both detection and situational awareness during outbreaks of acute respiratory infections. In this work, we describe the methods by which automated text analyses of chest imaging reports can combine with structured EMR data to accurately identify outpatients with pneumonia (sensitivities of 58-75%, and PPV of 64-86%).

Full Text:



Online Journal of Public Health Informatics * ISSN 1947-2579 *