Environment, Epidemiology, & Etiology Surveillance & Analysis Toolkit (E3SAT)

In the aftermath of Desert Shield and Desert Storm, researchers have struggled with explaining the array of serious health impairing symptoms that have become collectively known as Gulf War Syndrome.  Approximately 30 percent of the 700,000 U.S. servicemen and servicewomen in the first Persian Gulf War have registered in the Gulf War Illness database complaining of these symptoms.  A key stumbling block in researching Gulf War Syndrome is the absence of a means for integrating relevant data—soldier medical records, environmental health and surveillance data, deployment data—and discovering and analyzing to discover patterns and correlations between deployment exposures and soldier signs, symptoms, and potential causes.

In the Environment, Epidemiology, and Etiology Surveillance and Analysis Toolkit (E3SAT) initiative, KBSI is designing and developing a tool suite that allows researchers to collect, integrate and data mine medical records, environmental exposures, and deployment locations.  The E3SAT technology addresses the military’s need for a more encompassing data view that links soldier deployment locations, durations, Military Occupation Specialty codes (MOS), environmental incidences and exposures, and soldier medical records, including current signs and symptoms.  This encompassing data view, coupled with the ability to analyze the collected data, will facilitate the scientific study of environmental, epidemiological and etiological factors, helping researchers address the environmental and biological concerns of military personnel.

E3SAT will implement an innovative data collection, integration, and analyses software interface that will supplement the standard data available from current sources with information retrieved from public sources on local and regional environmental patterns and with epidemiology information.  The tool takes a knowledge-based approach to the fusion of disparate information sources, structured, and unstructured.  Researchers will be able to seamlessly interact with the integrated analysis cube, without jeopardizing existing analysis workflows and without adding undue work burdens.  It will allow for a time phased longitudinal view of individual records as well as latitudinal views across multiple soldiers.  Users can drill down into data supporting a discovered hypothesis, epidemiology, or pattern;  in addition, users can also drill up to view similar symptoms or environmental exposure clusters.  The E3SAT technology’s point and click interface will allows researchers to select a keyword, phrase, anomaly, or finding and traverse to related parts of knowledge bases and data records.  E3SAT will provide real-time Point of Care information, allowing clinicians to scan for issues in specific areas and to link patient signs and symptoms to similar clusters.  In addition, E3SAT will accept inputs and annotations from Subject Matter Experts (SME) and will log hypotheses from these SMEs for validation.

The E3SAT toolkit will also act as a diagnostic decision aid, red flagging anomalies and patterns that help identify the onset or origin of disease outbreaks or that indicate precipitating factors for health status changes.  As part of the E3SAT initiative, KBSI will be designing, testing, and validating techniques and templates that analysts can customize and apply for environmental, epidemiological, and etiological analysis.  Theses analysis techniques and algorithms will support pattern and anomaly discovery, correlation analysis, hypothesis testing, multi-variate analysis, odds ratio analysis, and association and temporal pattern discovery.

E3SAT will provide researchers and medical practitioners with a comprehensive means for collecting, analyzing, and annotating environmental, epidemiological, and etiological data concerning soldier health.  This capability benefits both the retrospective analysis of historical events (e.g., Gulf War syndrome) and a real-time analysis of current data for the detection of health anomalies and the potential outbreak of epidemics.  The E3SAT technology will contribute significantly to the nation’s environmental and epidemiological surveillance and analysis.

Phase II Developments

The E3SAT Phase II initiative is building off of the Phase I successes and focusing on technology transition.  As part of this work, KBSI is extending the data handling and data analysis capabilities, including developing the distributed data access and computational features, and extending the data mining algorithms and analysis capabilities.  KBSI will also integrate distributed data processing and data mining capabilities such as Globus, Hadoop, Weka4WS into E3SAT.

At the close of Phase II, KBSI will deploy a version of E3SAT that provides the data analysis infrastructure, models and templates to facilitate environmental, epidemiological and etiological analysis.  This version will support homeland security, DoD, DHHS, AABB, epidemiologists, and other public health officials in accessing and mining disparate data sources for their analyses.  The version will also provide analytical support to the international donor biovigilance community portal (IDBP), enabling blood transfusion specialists (www.isbt-web.org) to communicate, collaborate, discuss issues, analyze reactions, and identify interventions that will help improve donor safety.