Efficient Analytics for Effective Monitoring of Biomedical Security

Abstract

This paper reviews three successful statistical data mining approaches developed recently at the Auton Lab of Carnegie Mellon University to support public health officials in their work towards protecting biomedical safety and security. The presented methods focus on monitoring health care data sources including hospital emergency department records, sales of over-the-counter medications, and consumer food complaints. Their purpose is to detect statistically significant signs of disease outbreaks, or food safety related concerns, as early as possible. These approaches have already been successfully deployed in the United States and other developed countries, but they also have a vast potential utility among developing societies. The Auton Lab is actively seeking additional deployments, and several pieces of the relevant software are available for download and use free of charge. This paper describes each of the presented methods, and provides results of their utilization so far.