Bio-threat early warning system moves to next stage
The National Collaborative for Bio-Preparedness, which monitors a vast array of data from multiple sources using real-time analytics, is in the final stages of field testing.
An innovative bio-threat early warning system that uses analytics to enhance public safety has received $3 million in additional funding from the Homeland Security Department for further field testing and development of the project into an operational system.
The National Collaborative for Bio-Preparedness is designed to monitor emergency medical services reports, poison center data and a vast array of other data sets, including social media, to detect signs of biological threats.
The initiative gathers multiple sources of data – from human, animal and environmental sources –and employs advanced analytics to give emergency responders and health professionals the information they need to prevent the spread of bio-threats. During the demonstration phase, researchers, looking at EMS records, were able to detect a gastrointestinal outbreak a full two months before it was recognized by standard reporting.
NCB-Prepared is in the final stages of field testing by its primary partners, the University of North Carolina at Chapel Hill, North Carolina State University and the SAS Institute. Congressman David Price (D-N.C.), the ranking member of the House Homeland Security Appropriations Subcommittee, recently announced the additional funding for the system, which pools data from systems that currently do not talk to each other. NCB-Prepared was launched in 2010 with an initial $5 million installment from DHS’ Office of Health Affairs under a cooperative agreement.
Over the past 10 months, the NCB-Prepared team has worked to automate tasks and assess the timeliness, completeness and quality of the data being collected, said Dr. Charles Cairns, chair of emergency medicine at UNC and principal investigator at NCB-Prepared. This work involved testing the system to ensure that analytics could be performed on the data in real-time and that the system is providing valuable information about health threats, bio-symptoms and syndromes. The team is also testing to determine how the system works across a defined geographical area and if it can scale up from city to county to region to state levels, Cairns noted.
NCB-Prepared got feedback from public health officials, first responders and emergency management officials from local, state and federal governments on the system’s data quality and analytic and visualization approaches.
During this next grant period, the partners will enhance operations and determine how the system can be used by the federal government, especially DHS. “We are trying to make that transition into an operating system ready for optimization within the federal and, hopefully, state and local infrastructures,” Cairns said. Ongoing field tests involve seeding the system with data for two threat scenarios: influenza and food contamination.
The NCB-Prepared analytics system initially deployed SAS text analytics software running on NC State’s cloud-based Virtual Computing Lab to scan rapidly expanding data sets for patterns that could indicate an emerging threat to public health. Now, as part of the operational system, the team is running analytics on the SAS On-Demand cloud solution.
“We wanted to have a system that is extensible and could handle increasingly large data sets,” Cairns said. Additionally, reliability is critical, so a system that can run non-stop is essential. “We now get live data feeds that keep increasing from multiple data sets.” NCB-Prepared has collected more than 13 million records and continues to add data from national sources with plans to include information from more states.