Artificial Intelligence Early Warning System Installed at the Olympics For Bioterrorism Surveillance

February 13, 2002 | Source: KurzweilAI

An artificial intelligence computer system that analyzes state-wide patient data from emergency rooms and instant care facilities has been installed in most of the state of Utah for the Olympics. If it detects a significant pattern suggesting an outbreak, it pages the on-call state public health physician.
The Realtime Outbreak and Disease Surveillance (RODS), developed by the University of Pittsburgh and Carnegie Mellon University, is a protoype system that collects and analyzes relevant data automatically and in real-time, including emergency room registration data, microbiology culture results, reports of radiographs, and laboratory orders.

It uses several AI techniques for machine learning, natural language and data mining, including case definitions, automatic detection algorithms that can be attached to specific data streams, and data analytic tools that support temporal and spatial data analysis and visualization.

RODS was initially installed in Western Pennsylvania in August of 1999 for public health surveillance for the 3 million residents of 13 counties. So far, the system has detected only a naturally occurring outbreak of influenza. It is now being extended to all of Pennsylvania. Other public health departments in the U.S. are now showing interest in RODS. Given how quickly it was installed in Utah, RODS principal investigator Michael Wagner, M.D., believes it could also be implemented in other states fairly quickly.

Research funding has come from DARPA (Defense Advanced Research Projects Agency), the Centers for Disease Control and Prevention, the National Library of Medicine, and the Agency for Healthcare Research and Quality.