Authors: Sylvain DeLisle, Brett South, Jill A. Anthony, Ericka Kalp, Adi Gundlapallli, Frank C. Curriero, Greg E. Glass, Matthew Samore, Trish M. Perl
DOI: 10.1371/journal.pone.0013377
Abstract Summary
Researchers developed algorithms combining structured electronic medical record data (diagnostic codes, vital signs, prescriptions) with automated analysis of clinical notes to detect acute respiratory infections. The best approach achieved 69-73% sensitivity with 52-68% precision, significantly improving epidemic surveillance capabilities. This demonstrates how repurposed EMR data could enable earlier outbreak detection.
Why Brain? 🧠
Study shows combining structured EMR data with free-text analysis detects respiratory infections with up to 99% sensitivity, potentially enabling earlier epidemic detection and prevention of mass casualties.
The image is AI-generated for illustrative purposes only. Courtesy of Midjourney.
