Fuzzy association rule mining and classification for the prediction of malaria in South Korea.

BACKGROUND: Malaria is the world's most prevalent vector-borne disease. Accurate prediction of malaria outbreaks may lead to public health interventions that mitigate disease morbidity and mortality. METHODS: We describe an application of a method for creating prediction models utilizing Fuzzy Association Rule Mining to extract relationships between epidemiological, meteorological, climatic, and socio-economic data from Korea. These relationships are in the form of rules, from which the best set of rules is automatically chosen and forms a classifier. Two classifiers have been built and their results fused to become a malaria prediction model. Future malaria cases are predicted as Low, Medium or High, where these classes are defined as a total of 0-2, 3-16, and above 17 cases, respectively, for a region in South Korea during a two-week period. Based on user recommendations, HIGH is considered an outbreak. RESULTS: Model accuracy is described by Positive Predictive Value (PPV), Sensitivity, and F-score for each class, computed on test data not previously used to develop the model. For predictions made 7-8 weeks in advance, model PPV and Sensitivity are 0.842 and 0.681, respectively, for the HIGH classes. The F0.5 and F3 scores (which combine PPV and Sensitivity) are 0.804 and 0.694, respectively, for the HIGH classes. The overall FARM results (as measured by F-scores) are significantly better than those obtained by Decision Tree, Random Forest, Support Vector Machine, and Holt-Winters methods for the HIGH class. For the Medium class, Random Forest and FARM obtain comparable results, with FARM being better at F0.5, and Random Forest obtaining a higher F3. CONCLUSIONS: A previously described method for creating disease prediction models has been modified and extended to build models for predicting malaria. In addition, some new input variables were used, including indicators of intervention measures. The South Korea malaria prediction models predict Low, Medium or High cases 7-8 weeks in the future. This paper demonstrates that our data driven approach can be used for the prediction of different diseases.

Prediction of high incidence of dengue in the Philippines.

BACKGROUND: Accurate prediction of dengue incidence levels weeks in advance of an outbreak may reduce the morbidity and mortality associated with this neglected disease. Therefore, models were developed to predict high and low dengue incidence in order to provide timely forewarnings in the Philippines. METHODS: Model inputs were chosen based on studies indicating variables that may impact dengue incidence. The method first uses Fuzzy Association Rule Mining techniques to extract association rules from these historical epidemiological, environmental, and socio-economic data, as well as climate data indicating future weather patterns. Selection criteria were used to choose a subset of these rules for a classifier, thereby generating a Prediction Model. The models predicted high or low incidence of dengue in a Philippines province four weeks in advance. The threshold between high and low was determined relative to historical incidence data. PRINCIPAL FINDINGS: Model accuracy is described by Positive Predictive Value (PPV), Negative Predictive Value (NPV), Sensitivity, and Specificity computed on test data not previously used to develop the model. Selecting a model using the F0.5 measure, which gives PPV more importance than Sensitivity, gave these results: PPV = 0.780, NPV = 0.938, Sensitivity = 0.547, Specificity = 0.978. Using the F3 measure, which gives Sensitivity more importance than PPV, the selected model had PPV = 0.778, NPV = 0.948, Sensitivity = 0.627, Specificity = 0.974. The decision as to which model has greater utility depends on how the predictions will be used in a particular situation. CONCLUSIONS: This method builds prediction models for future dengue incidence in the Philippines and is capable of being modified for use in different situations; for diseases other than dengue; and for regions beyond the Philippines. The Philippines dengue prediction models predicted high or low incidence of dengue four weeks in advance of an outbreak with high accuracy, as measured by PPV, NPV, Sensitivity, and Specificity.

Developing open source, self-contained disease surveillance software applications for use in resource-limited settings.

BACKGROUND: Emerging public health threats often originate in resource-limited countries. In recognition of this fact, the World Health Organization issued revised International Health Regulations in 2005, which call for significantly increased reporting and response capabilities for all signatory nations. Electronic biosurveillance systems can improve the timeliness of public health data collection, aid in the early detection of and response to disease outbreaks, and enhance situational awareness. METHODS: As components of its Suite for Automated Global bioSurveillance (SAGES) program, The Johns Hopkins University Applied Physics Laboratory developed two open-source, electronic biosurveillance systems for use in resource-limited settings. OpenESSENCE provides web-based data entry, analysis, and reporting. ESSENCE Desktop Edition provides similar capabilities for settings without internet access. Both systems may be configured to collect data using locally available cell phone technologies. RESULTS: ESSENCE Desktop Edition has been deployed for two years in the Republic of the Philippines. Local health clinics have rapidly adopted the new technology to provide daily reporting, thus eliminating the two-to-three week data lag of the previous paper-based system. CONCLUSIONS: OpenESSENCE and ESSENCE Desktop Edition are two open-source software products with the capability of significantly improving disease surveillance in a wide range of resource-limited settings. These products, and other emerging surveillance technologies, can assist resource-limited countries compliance with the revised International Health Regulations.

The collaborative experience of creating the National Capital Region Disease Surveillance Network.

The Johns Hopkins University Applied Physics Laboratory (JHU/APL) implemented state and district surveillance nodes in a central aggregated node in the National Capital Region (NCR). Within this network, de-identified health information is integrated with other indicator data and is made available to local and state health departments for enhanced disease surveillance. Aggregated data made available to the central node enable public health practitioners to observe abnormal behavior of health indicators spanning jurisdictions and view geographical spread of outbreaks across regions.Forming a steering committee, the NCR Enhanced Surveillance Operating Group (ESOG), was key to overcoming several data-sharing issues. The committee was composed of epidemiologists and key public health practitioners from the 3 jurisdictions. The ESOG facilitated early system development and signing of the cross-jurisdictional data-sharing agreement. This agreement was the first of its kind at the time and provided the legal foundation for sharing aggregated health information across state/district boundaries for electronic disease surveillance.Electronic surveillance system for the early notification of community-based epidemics provides NCR users with a comprehensive regional view to ascertain the spread of disease, estimate resource needs, and implement control measures. This article aims to describe the creation of the NCR Disease Surveillance Network as an exceptional example of cooperation and potential that exists for regional surveillance activities.

Utility of the ESSENCE Surveillance System in Monitoring the H1N1 Outbreak.

The Electronic Surveillance System for the Early Notification of Community-Based Epidemics (ESSENCE) enables health care practitioners to detect and monitor health indicators of public health importance. ESSENCE is used by public health departments in the National Capital Region (NCR); a cross-jurisdictional data sharing agreement has allowed cooperative health information sharing in the region since 2004. Emergency department visits for influenza-like illness (ILI) in the NCR from 2008 are compared to those of 2009. Important differences in the rates, timing, and demographic composition of ILI visits were found. By monitoring a regional surveillance system, public health practitioners had an increased ability to understand the magnitude and character of different ILI outbreaks. This increased ability provided crucial community-level information on which to base response and control measures for the novel 2009 H1N1 influenza outbreak. This report underscores the utility of automated surveillance systems in monitoring community-based outbreaks. There are several limitations in this study that are inherent with syndrome-based surveillance, including utilizing chief complaints versus confirmed laboratory data, discerning real disease versus those healthcare-seeking behaviors driven by panic, and reliance on visit counts versus visit rates.