skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Biosurveillance Using Clinical Diagnoses and Social Media Indicators in Military Populations

Abstract

U.S. military influenza surveillance uses electronic reporting of clinical diagnoses to monitor health of military personnel and detect naturally occurring and bioterrorism-related epidemics. While accurate, these systems lack in timeliness. More recently, researchers have used novel data sources to detect influenza in real time and capture nontraditional populations. With data-mining techniques, military social media users are identified and influenza-related discourse is integrated along with medical data into a comprehensive disease model. By leveraging heterogeneous data streams and developing dashboard biosurveillance analytics, the researchers hope to increase the speed at which outbreaks are detected and provide accurate disease forecasting among military personnel.

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1]
  1. Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1345453
Report Number(s):
PNNL-26271
400403909
DOE Contract Number:
AC05-76RL01830
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEFENSE; 59 BASIC BIOLOGICAL SCIENCES; biosurveillance; social media; biodefense; military

Citation Formats

Corley, Courtney D., Volkova, Svitlana, Rounds, Jeremiah, Charles-Smith, Lauren E., Harrison, Joshua J., Mendoza, Joshua A., and Han, Keith S.. Biosurveillance Using Clinical Diagnoses and Social Media Indicators in Military Populations. United States: N. p., 2017. Web. doi:10.2172/1345453.
Corley, Courtney D., Volkova, Svitlana, Rounds, Jeremiah, Charles-Smith, Lauren E., Harrison, Joshua J., Mendoza, Joshua A., & Han, Keith S.. Biosurveillance Using Clinical Diagnoses and Social Media Indicators in Military Populations. United States. doi:10.2172/1345453.
Corley, Courtney D., Volkova, Svitlana, Rounds, Jeremiah, Charles-Smith, Lauren E., Harrison, Joshua J., Mendoza, Joshua A., and Han, Keith S.. Thu . "Biosurveillance Using Clinical Diagnoses and Social Media Indicators in Military Populations". United States. doi:10.2172/1345453. https://www.osti.gov/servlets/purl/1345453.
@article{osti_1345453,
title = {Biosurveillance Using Clinical Diagnoses and Social Media Indicators in Military Populations},
author = {Corley, Courtney D. and Volkova, Svitlana and Rounds, Jeremiah and Charles-Smith, Lauren E. and Harrison, Joshua J. and Mendoza, Joshua A. and Han, Keith S.},
abstractNote = {U.S. military influenza surveillance uses electronic reporting of clinical diagnoses to monitor health of military personnel and detect naturally occurring and bioterrorism-related epidemics. While accurate, these systems lack in timeliness. More recently, researchers have used novel data sources to detect influenza in real time and capture nontraditional populations. With data-mining techniques, military social media users are identified and influenza-related discourse is integrated along with medical data into a comprehensive disease model. By leveraging heterogeneous data streams and developing dashboard biosurveillance analytics, the researchers hope to increase the speed at which outbreaks are detected and provide accurate disease forecasting among military personnel.},
doi = {10.2172/1345453},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Feb 23 00:00:00 EST 2017},
month = {Thu Feb 23 00:00:00 EST 2017}
}

Technical Report:

Save / Share:
  • This work is the first to take advantage of recurrent neural networks to predict influenza-like-illness (ILI) dynamics from various linguistic signals extracted from social media data. Unlike other approaches that rely on timeseries analysis of historical ILI data [1, 2] and the state-of-the-art machine learning models [3, 4], we build and evaluate the predictive power of Long Short Term Memory (LSTMs) architectures capable of nowcasting (predicting in \real-time") and forecasting (predicting the future) ILI dynamics in the 2011 { 2014 influenza seasons. To build our models we integrate information people post in social media e.g., topics, stylistic and syntactic patterns,more » emotions and opinions, and communication behavior. We then quantitatively evaluate the predictive power of different social media signals and contrast the performance of the-state-of-the-art regression models with neural networks. Finally, we combine ILI and social media signals to build joint neural network models for ILI dynamics prediction. Unlike the majority of the existing work, we specifically focus on developing models for local rather than national ILI surveillance [1], specifically for military rather than general populations [3] in 26 U.S. and six international locations. Our approach demonstrates several advantages: (a) Neural network models learned from social media data yield the best performance compared to previously used regression models. (b) Previously under-explored language and communication behavior features are more predictive of ILI dynamics than syntactic and stylistic signals expressed in social media. (c) Neural network models learned exclusively from social media signals yield comparable or better performance to the models learned from ILI historical data, thus, signals from social media can be potentially used to accurately forecast ILI dynamics for the regions where ILI historical data is not available. (d) Neural network models learned from combined ILI and social media signals significantly outperform models that rely solely on ILI historical data, which adds to a great potential of alternative public sources for ILI dynamics prediction. (e) Location-specific models outperform previously used location-independent models e.g., U.S. only. (f) Prediction results significantly vary across geolocations depending on the amount of social media data available and ILI activity patterns.« less
  • The scope of environmental impact statements prepared during the past few years has steadily expanded to incorporate all aspects of the social as well as the natural environment, including demographic, economic, social, political, and cultural conditions. Broadly conceived, social impacts are alterations in people's living conditions that occur in conjunction with a new policy, program, or project, and that (1) are in addition to all other concurrent changes produced by other factors, and (2) are seen by those affected as significant social events. Since any social environment is constantly changing, the crucial problems in analyzing social impacts are to identifymore » those social alterations that are a direct or indirect result of the specific action under examination, apart from all other events and changes, and to determine which of these alterations are having significant social effects on the people involved. Three features of this conception of social impacts are especially noteworthy. First, although impacts are often thought of as undesirable or detrimental in nature, they may also be desirable or beneficial. Second, although impacts are often described as caused by prior intervening innovations, in reality they always interact with their original causes in a reciprocal process, either immediately or after some time lag. Third, the purpose of social impact assessment is to enable policy makers to anticipate and plan for potential impacts before they occur, and then act to prevent or mitigate undesired impacts. A new methodology for performing social impact assessment and management studies that meet current needs by emphasizing standardized social indicators and social planning techniques is proposed. We refer to our approach as the Social Impact and Planning (SIP) method of social impact assessment. (ERB)« less
  • The methodology proposed essentially argues that mankind is inexorably dependent on the natural environment, and that collective social life is always constrained and shaped by the basic ecological factors of natural resources, population characteristics, material and social technology, and the economic order which satisfies peoples' sustenance needs. To this ecological perspective two further assumptions are added: (1) collective social activities are generally aimed at the attainment of goals that reflect the values and interests of the participants, so that social impact assessment must reflect both the ecological conditions prevailing in a community and the values, interests, and goals of themore » community members; and (2) a community can be viewed as a problem-solving social system, in which (a) challenges such as new environmental or technological conditions initially disrupt existing social processes and patterns of social ordering, which (b) creates temporary problem conditions and activities, which in turn (c) generate collective responses to cope with these problems, which finally (d) act back (as either positive or negative feedback) on the initial disruptive conditions. The inputs to a general social impact model are (a) such disruptive innovations or alterations as growing resource scarcities (e.g., oil or water depletion), governmental policy decisions (e.g., emphasis on solar energy, limiting economic growth), or technological development projects (e.g., nuclear energy centers, water reservoirs); and (b) values and interests of the people in the affected area. Whatever the precise nature of the outside disruption, it can be expected to have direct and relatively immediate impacts on the basic ecological factors of the local population and economy. (MCW)« less
  • Social media can provide a resource for characterizing communities and small populations through activities and content shared online. For instance, studying the language use in social media within military populations may provide insights into their health and wellbeing. In this paper, we address three research questions: (1) How do military populations use social media? (2) What do military users discuss in social media? And (3) Do military users talk about health and well-being differently than civilians? Military Twitter users were identified through keywords in the profile description of users who posted geo-tagged tweets at military installations. The data was anonymizedmore » for the analysis. User profiles that belong to military population were compared to the nonmilitary population. Our results indicate that military users talk more about events in their military life, whereas nonmilitary users talk more about school, work, and leisure activities. We also found that the online content generated by both populations is significantly different, including health-related language and communication behavior.« less