Skip to main content

Displaying results 401 - 408 of 578
  • Fast Graph Structure Learning from Unlabeled Data for Outbreak Detection

    Content Type: Abstract

    Disease surveillance data often has an underlying network structure (e.g. for outbreaks which spread by person-to-person contact). If the underlying graph structure is known, detection methods such as GraphScan (1) can be used to identify an… read more
    … connected subgraph for each graph structure and each training example using GraphS- can. We normalize each score … by the maximum unconstrained subset score for that training example (computed efficiently using LTSS). We then compute the mean normalized score averaged over all training examples. If a given graph is close to the true …
  • Results from the BioSense Jurisdiction-Specific Wbinars

    Content Type: Abstract

    BioSense is a Centers for Disease Control and Prevention (CDC) national near real-time public health surveillance system. CDC’s BioIntelligence Center (BIC) analysts monitor, analyze, and interpret BioSense data daily and provide support to BioSense… read more
    … hospital utilization and mortality data. Identified training needs included the following: 1) how to use the … up during an event, and 2) self- paced, interactive training materials and tools that will enable users to … this level of dialogue, as well as develop additional training tools to provide ongoing support for our users. …
  • Building a Community of Practice

    Content Type: Webinar

    Presented September 27, 2018. This presentation offers practical tips on how to create a successful and sustainable community of practice. Presenter Deborah W. Gould, PhD., Division of Health Informatics and Surveillance,… read more
    … https://www.dau.mil/sites/governance-and-training/SitePages/Guide to Establishing Communities.aspx … https://www.dau.mil/sites/governance-and-training/SitePages/Guide to Establishing Communities.aspx …
  • Anomaly Pattern Detection for Biosurveillance

    Content Type: Abstract

    We propose a new method for detecting patterns of disease cases that correspond to emerging outbreaks. Our Anomaly Pattern Detector (APD) first uses a "local anomaly detector" to identify individually anomalous records and then searches over subsets… read more
    … particular rules for the current (test) and historical (training) datasets. How- ever, an outbreak may create a … we compare it to the corres- ponding subset in the training data. For each rule R, we determine the total number of corresponding records in the test and training datasets (C(R)test and C(R)train) and the number of …
  • Evaluation of Preprocessing Techniques for Chief Complaint Classification

    Content Type: Abstract

    The Real-time Outbreak and Disease Surveillance system collects chief complaints as free text and uses a naïve Bayesian classifier called CoCo to classify the complaints into syndromic categories. CoCo 3.0 has been trained on 28,990 manually clas-… read more
    … 10,161 chief complaints not previously involved in CoCo’s training to measure the propor- tion of chief complaints … We counted the number of unique words in the train- ing set for CoCo 3.0 prior to and post preprocessing, … plaints and decreased the number of unique words in the training set from 2,775 to 2,308. All the words changed in …
  • Infection Control Practitioner Use of NC DETECT

    Content Type: Abstract

    The UNC Department of Emergency Medicine (UNC DEM) conducted an online survey to better understand the surveillance needs of Infection Control Practitioners (ICPs) in North Carolina and solicit feedback on the utility of the North Carolina Disease… read more
    … of the reports available to ICPs and provide targeted training on the specific surveillance needs of ICPs. In … Program for Infection Control and Epidemiology (SPICE). Training should focus on the tools in NC DETECT that assist …
  • Learning Specific Detectors of Adverse Events in Multivariate Time Series

    Content Type: Abstract

    This paper describes how powerful detectors of adverse events manifested in multivariate series of bio-surveillance data can be learned using only a few labeled instances of such events.
    … using domain expertise, if the amount of available training data is insufficient to support automated learning … for machine learning techniques which would allow for training specific detectors even if the number of iden- … improvement can be obtained by combining into the training data labels on false posi- tives with one, then …
  • Enabling User-Driven Public Health Analyses through Automated Data Querying

    Content Type: Abstract

    Public health officials are now receiving more data than ever in electronic formats, and also stand to benefit more than ever from ongoing advances in the medical and epidemiological sciences. At the same time, this growing … read more
    … for the Early Notification of Community-Based Epidemics (ESSENCE II).” Journal of Urban Health, Volume 80(2)S1: …