Understanding syndromic hotspots - a visual analytics approach

Abstract

When analyzing syndromic surveillance data, health care officials look for areas with unusually high cases of syndromes. Unfortunately, many outbreaks are difficult to detect because their signal is obscured by the statistical noise. Consequently, many detection algorithms have a high false positive rate. While many false alerts can be easily filtered by trained epidemiologists, others require health officials to drill down into the data, analyzing specific segments of the population and historical trends over time and space. Furthermore, the ability to accurately recognize meaningful patterns in the data becomes more challenging as these data sources increase in volume and complexity. To facilitate more accurate and efficient event detection, we have created a visual analytics tool that provides analysts with linked geo-spatiotemporal and statistical analytic views. We model syndromic hotspots by applying a kernel density estimation on the population sample. When an analyst selects a syndromic hotspot, temporal statistical graphs of the hotspot are created. Similarly, regions in the statistical plots may be selected to generate geospatial features specific to the current time period. Demographic filtering can then be combined to determine if certain populations are more affected than others. These tools allow analysts to perform real-time hypothesis testing and evaluation. MOTIVATIONRecently, the detection of adverse health events has focused on pre-diagnosis information to improve response time. This type of detection is more largely termed syndromic surveillance and involves the collection and analysis of statistical health trend data, most notably symptoms reported by individuals seeking care in emergency departments. Currently, the Indiana State Department of Health (ISDH) employs a state syndromic surveillance system called PHESS (Public Health Emergency Surveillance System) [9], which receives electronically transmitted patient data (in the form of emergency department chief complaints) from 73 hospitals around the state at an average rate of 7000 records per day.These complaints are then classified into nine categories (respiratory, gastro-intestinal, hemorrhagic, rash, fever, neurological, botulinic, shock/coma, and other) [4] and used as indicators to detect public health emergencies before such an event is confirmed by diagnoses or overt activity. Unfortunately, detection of events from these indicators is an extremely challenging issue. Figure 1 shows a typical month of emergency department visits for those complaints classified as neurological syndromes. During this time period, there was one event of carbon monoxide poisoning which happened to coincide with the largest peak on December 21st; however, this peak is not significantly higher than any other peak during this month. Obviously, the detection of such a small signal deviation can be extremely difficult. In order to facilitate enhanced syndromic surveillance and improve signal detection, we have developed a linked geospatiotemporal vi...

A Visual Analytics Approach to Understanding Spatiotemporal Hotspots

“…We have also extended our previous system [31] to include the spatiotemporal history, bivariate and multivariate exploration, thresholding and color mapping tools. Contour histories provide users with geo-spatiotemporal views of current and past data trends, allowing them to track hotspot movement across time, or look for correlations between multiple variables.…”

“…In order to facilitate enhanced data exploration and improve signal detection, we have developed a linked geo-spatiotemporal visual analytics tool designed for advanced data exploration. This paper presents a set of extensions to our previous suite of visual analytics tools [31] for the enhanced exploration of multivariate geospatiotemporal data. Our system features include:…”

An Integrated In-Situ Approach to Impacts from Natural Disasters on Critical Infrastructures

“…This comprehensive environment requires integrated visual and traditional systematic analysis of massive data, including improved strategies for exploratory visual analysis, hypothesis testing and user-specific presentation of relevant information as a basis for actionable decision making. Furthermore, visual analytics tools for analyzing syndromic hotspots are presented by Maciejewski et al [12] and allow the analyst to perform real-time hypothesis testing and evaluation.…”

Using Topological Analysis to Support Event-Guided Exploration in Urban Data

“…2). Maciejewski et al [37] used kernel density estimation to model hotspots in spatial distributions along with time series analysis to detect anomalous hotspots. Andrienko et al [3,4,5,6] proposed visual analytics procedures to determine places of interest based on high-frequency events that also have high density of occurrence.…”