INNOVATION IN OBSERVATION: EARLY OUTBREAK DETECTION USING REMOTE SENSING

Health threats stemming from environmental contamination or purposeful acts of bioterrorism call for a worldwide effort in improving early outbreak detection, with the goal of ameliorating current and future risks. When a single case of a disease of great concern constitutes an outbreak, the problem of outbreak detection is straightforward and easy. However, for the vast majority of maladies, a single data source is available and a simple analytical solution does not exist. Furthermore, each step in developing reliable, sensitive, effective surveillance systems demonstrates enormous complexities in the transmission, manifestation, detection, and control of emerging health threats. In this communication, potential future innovations in early outbreak detection systems that can overcome the pitfalls of current surveillance, including the use of remotely sensed data, will be examined. Special attention should be paid to spatial and temporal alignments of data abstracted from various streams and sources, the nature of delayed and time-distributed effects, understanding the effects of missingness patterns on our ability to assess associations, and strategies for improving the quality of data curation and verification. We believe that modern advances in assembling data, techniques for collating and processing information, and technology that enables integrated analysis will facilitate a new paradigm in outbreak definition, detection, and evaluation. We anticipate that moving forward in this direction will provide the highly desired sensitivity and specificity to meet the emerging challenges of global disease surveillance.