Skip to main content

Displaying results 1 - 8 of 9
  • in silico Surveillance: Using Detailed Computer Simulations to Develop and Evaluate Outbreak Detection

    Content Type: Abstract

    Developing and evaluating outbreak detection is challenging for many reasons.  A central difficulty is that the data the detection algorithms are “trained” on are often relatively short historical samples and thus do not represent the full… read more
    … variable results even under identical initial conditions. Due to their ability to mimic a wide range outcomes and more … results even under identi- cal initial conditions. Due to their ability to mimic a wide range outcomes and more … The individuals, locations, and activity patterns are all based on data collected from the real world and are …
  • in silico Surveillance: Highly detailed agent-based models for surveillance system evaluation and design

    Content Type: Abstract

    Modern public health surveillance systems have great potential for improving public health. However, evaluating the performance of surveillance systems is challenging because examples of baseline disease distribution in the population are limited to… read more
    … and complete history of disease progression for all individuals in the population; those who are in a … the different surveillance systems. Increased coverage did not linearly improve detection probability for all … systems with uniform coverage of the population did not perform better than one based on a real-world system …
  • Empirical/Asymptotic P-Values for Monte Carlo-Based Hypothesis Testing: An Application to Cluster Detection Using the Scan Statistic

    Content Type: Abstract

    SaTScan is a freely available software that uses the scan statistic to detect clusters in space, time or space-time. SaTScan uses Monte Carlo hypothesis testing in order to produce a p-value for the null hypothesis that no… read more
    … 999 replicates the maximum log- likelihood ratio, among all distinct circles, is the sta- tistic reported. The … including the Gamma, Normal, and Lognormal, all resulted in biased p-values. Interest- ingly, the … Communications in Sta- tistics: Theory and Methods, 1997. 26: p. 1481-1496. [3] Gumbel, EJ, Statistics of Extremes. …
  • Nurse Call Data for Detection of Influenza-Like Illness

    Content Type: Abstract

    Influenza affects millions of people and causes about 36,000 deaths in the United States each winter. Pandemics of influenza emerge at irregular intervals. National influenza surveillance is used to detect the emergence and spread of influenza virus… read more
    … 3 views Submitted by elamb on Tue, 03/26/2019 - 16:22 … isolates, percentage ILI, or both indica- tors for all regions. The peak week for calls fell within one week of … illness in ambulatory care setting. erg Infect Dis 2004;10:1806-11. ] Ritzwoller DP, Kleinman K, Palen TE, …
  • Multi-Syndrome Analysis of Time Series: A New Concept for Outbreak Investigation

    Content Type: Abstract

    Temporal anomaly detection is a key component of real time surveillance. Today, despite the abundance of temporal information on multiple syndromes, multivariate investigation of temporal anomalies remains under-explored. Traditionally, an outbreak… read more
    … the daily syndromic rates into the five dimensional space de- fined by the eigenvectors of the correlation matrix of … Thus, this framework can detect outlying signals that are due to mean shifts, scale shifts, or both. Part of the data …
  • Clustering under the Null: One Reason for Too Many Signals

    Content Type: Abstract

    I examine the nature and expression of the null hypothesis often used in spatial surveillance. I also show an example of how incorrect specification of the null can lead to excess signals without interesting outbreaks, and argue that this may be a… read more
    … a binomial distribution within each region, where all regions share a binomial probability p. This is often … zip codes in eastern Massachusetts under 1) a binomial dis- tribution with N = 23, p=.25 (the usual null) and 2) a … the mixture re- sulted in a p-value smaller than .05 on all but one of 386 days using both the space-time …
  • Evaluating Outbreak-Detection Methods Using Simulations: Volume Under the Time-ROC Surface

    Content Type: Abstract

    There are many proposed methods of identifying outbreaks of disease in surveillance data. However, there is little agreement about appropriate ways to choose amongst them. One common basis for comparison is simulating outbreaks and adding the simu… read more
    … of 1 suggests perfect sensitivity and specificity, with all the signals generated at the first possible moment. A … can be weighted to ascribe more importance to early de- tection. To avoid notation, we provide only a heuristic de- scription of each method. 1) The ‘tent’ method be- gins …
  • Multi-Method Comparison of Detecting Common Events of Public Health Interest: a Multi-Site, Multi-Stream Simulation Study

    Content Type: Abstract

    Existing statistical methods can perform well in detecting simulated bioterrorism events. However, these methods have not been well-evaluated for detection of the type of respiratory and gastrointestinal events of greatest… read more
    … Existing statistical methods can perform well in de- tecting simulated bioterrorism events [1]. However, these methods have not been well-evaluated for de- tection of the type of respiratory and gastrointestinal … signal extent was nearly as good as the best method in all cases. Table 2: Results Sim. Best method WROC (max = 1) …