Techniques for Early Detection of Localized Exposure to an Agent Active on a Biological Population
There are significant limitations with previous attempts at constructing early warning bio-surveillance systems. Conventional bio-surveillance focuses on categorical data collected from emergency rooms, clinics, and other healthcare facilities. The detection algorithms in these conventional systems rely on threshold crossing algorithms applied to single streams of data. Such an approach does not make optimal use of available information and cannot detect a bio-terrorist attack until sizeable numbers of infected individuals appear at healthcare facilities.
Techniques are provided for early detection of localized exposure to an agent active on a biological population. The techniques include collecting time series for each data type of multiple different data types. The data types are relevant for detecting exposure to the agent. For each data type, multiple time series are collected for corresponding multiple locations associated with the data type. Measures of anomalous conditions are generated at the locations for each of the different data types. The measures of anomalous conditions are based on the time series and a temporal model for each data type. Cluster analysis is performed on the measures of anomalous conditions to determine an estimated location, and an estimated extent, of effects from the agent. In various aspects, the techniques include a method, a computer-readable medium, and a system that implement the steps described above.
The techniques allow a surveillance system to more rapidly detect an event by combining signals spread over multiple data types with information about expected characteristics of the signal in those various data types. Furthermore, the techniques allow the surveillance system to avoid diluting the signal of a localized outbreak over too large an analysis area by focusing a detector on a spatial cluster identified by cluster analysis. In addition, the techniques allow the surveillance system to avoid consuming excessive resources in computing an exposure event in multiple source detectors, such as an exposure event associated with a best matched replica in a matched filter detector, by focusing the application of the multiple source detector in the vicinity of the cluster.
Patent Status: U.S. patent(s) 7266484 issued.
*JHU/APL is seeking an exclusive licensee and development partner for this technologyCONTACT: