Skip to main content

Automation

Description

Current practices of automated case detection fall into the extremes of diagnostic accuracy and timeliness. In regards to diagnostic accuracy, electronic laboratory reporting (ELR) is at one extreme and syndromic surveillance is at the other. In regards to timeliness, syndromic surveillance can be immediate, and ELR is delayed 7 days from initial patient visit. A plausible solution, a middle way, to the extremes of diagnostic precision and timeliness in current case detection practices is an automated Bayesian diagnostic system that uses all available data types, for example, freetext ED reports, radiology reports, and laboratory reports.We have built such a solution - Bayesian case detection (BCD). As a probabilistic system, BCD operates across the spectrum of diagnostic accuracy, that is, it outputs the degree of certainty for every diagnosis. In addition, BCD incorporates multiple data types as they appear during the course of a patient encounter or lifetime, with no degradation in the ability to perform diagnosis.

 

Objective

This paper describes the architecture and evaluation of our recently developed automated BCD system.

Submitted by hparton on
Description

Patient consultations recorded as voice dictations are frequently stored electronically as transcriptions in free text format. The information stored in free text is not computer tractable. Advances in artificial intelligence permit the conversion of free text into structured information that allows statistical analysis.

 

Objective

This paper describes DMReporter, a medical language processing system that automatically extracts information pertaining to diabetes (demography, numerical measurement values, medication list, and diagnoses) from the free text in physicians’ notes and stores it in a structured format in a MYSQL database.

Submitted by hparton on
Description

Public health departments have a strong interest in monitoring the incidence, care, and complications of gestational diabetes, as it is associated with poor outcomes for infants and increased risk of diabetes type II for mothers. Gestational diabetes rates are also a possible early marker for changes in the incidence of diabetes type II in the general population. However, diabetes is not generally a reportable condition and therefore, public health surveillance is limited to periodic telephone surveys (subject to self-report inaccuracies), sponsored clinical examinations (expensive, small sample size, no information about processes of care), and occasional research studies. Automated analysis of electronic health record data is a promising method to complement existing surveillance tools with longitudinal, continually updated, clinically rich data derived from large populations. We describe a pilot project to automatically survey electronic health record data in order to identify cases of gestational diabetes, describe their patterns of care and complications, and report summary data to the state health department.

 

Objective

To develop an electronic, prospective surveillance system to describe the incidence, care, and complications of gestational diabetes using live electronic health record data from a large defined population.

Submitted by hparton on
Description

Real-Time Biosurveillance Program (RTBP) introduces modern surveillance technology to health departments in Sri Lanka and Tamil Nadu, India. Triage data from each patient visit (basic demographics, signs, symptoms, preliminary diagnoses) is recorded on paper at health facilities. Case records are transmitted daily to a central database using the RTBP mobile phone application. It is done by medical professionals in India, but in Sri Lanka, due to staffing constraints, the same duty is performed by lower cost personnel with limited domain knowledge. That results in noticeable differences in data entry error rates between the two locations. Most of such issues are due to systematic and subjectivemisinterpretations of the handwritten doctor notes by the data entry personnel. If not identified and remedied quickly, these errors can adversely affect accuracy and timeliness of health events detection. There is a need to support system managers in their efforts to maintain high reliability of data used for public health surveillance.

 

Objective

We present a method for automated detection of systematic data entry errors in real time biosurveillance.

Submitted by hparton on
Description

Syndromic surveillance systems significantly enhance the ability of Public Health Units to identify, quantify, and respond to disease outbreaks. Existing systems provide excellent classification, identification, and alerting functions, but are limited in the range of statistical and mapping analyses that can be done. Currently available commercial off-the-shelf (COTS) statistical and GIS packages provide a much broader range of analytical and visualization tools, as well as the capacity for automation through user-friendly scripting languages. This study retrospectively evaluates the use of these packages for surveillance using syndromic data collected in Ottawa during the 2009 pH1NI outbreak.

 

Objective

The objective of this study was to create and evaluate a system that uses customized scripts developed for COTS statistical and GIS software to (1) analyze syndromic data and produce regular reports to public health epidemiologists, containing the information they would need to detect and manage an ILI outbreak, and (2) facilitate the generation more detailed analyses relevant to specific situations using these data.

Submitted by hparton on
Description

One objective of public health surveillance is detecting disease outbreaks by looking for changes in the disease occurrence, so that control measures can be implemented and the spread of disease minimized. For this purpose, the Florida Department of Health (FDOH) employs the Electronic Surveillance System for Early Notification of Community-based Epidemics (ESSENCE). The current problem was spawned by a laborintensive process at the FDOH: authentic outbreaks were detected by epidemiologists inspecting ESSENCE time series and derived event lists. The corresponding records indicated that patients arrived at an ED within a short interval, often less than 30minutes. The time-of-arrival (TOA) task was to develop and automate a capability to detect events with clustered patient arrival times at the hospital level for a list of subsyndrome categories of concern to the monitoring counties.

 

Objective

This presentation discusses the approach and results of collaboration to enable a solution of a hospital TOA monitoring problemin syndromic surveillance applied to public health data at the hospital level for county monitoring.

Submitted by hparton on
Description

In November of 2011, the local Public Health unit responsible for the Edmonton area (population 1.2mil) was alerted to an individual meeting the case definition for measles in the ED. A key part of the management strategy was to identify contacts to the index case, perform a risk assessment and, if applicable, inform them of the risk. Given the transmission characteristics, the risk for this group was defined as those present within the geographic area/environment of the index case within a specified time period. Public Health utilized the established manual lookup of hospital records and piloted an automated data query through the syndromic surveillance system, ARTSSN. This served as opportunity to validate the ability to generate a contact list, based on risk geography and time, of the ARTSSN system, and to compare the timeliness of each result.

Objective

Following a clinical case of measles presenting to an urban emergency department (ED), the local health authority sought to identify all patients that might be at risk for disease. This list of contacts was generated through a manual search of hospital records and through a piloted automated data query of the health authority's syndromic surveillance system, Alberta Real Time Syndromic Surveillance Net (ARTSSN). The purpose of this pilot study was to: 1) compare the completeness of the two lookup methods and, 2) describe the time requirements needed for each method.

Submitted by elamb on
Description

An increase in tuberculosis (TB) among homeless men residing in Marion County, Indiana was noticed in the summer of 2008. The Marion County Public Health Department (MCPHD) hosted screening events at homeless shelters in hopes of finding unidentified cases. To locate men who had a presumptive positive screen, the MCPHD partnered with researchers at Regenstrief Institute (RI) to create an alert for health care providers who use the Gopher patient management system in one of the city's busiest emergency departments. A similar process was used at this facility to impact prescription behavior.[1] A similar method was also used at the New York City Department of Health and Mental Hygiene.[2]

Submitted by elamb on
Description

Bioterrorism surveillance is an integral component of DCHD’s Comprehensive Emergency Management Plan. This study was a collaborative effort between Duval County Health Department, University of South Florida’s Center for Biological Defense (CBD), and DataSphere, LLC. DCHD’s role in the project was to identify surveillance sites, involve community partners, share data/info with surrounding agencies, counties and the state department of health, and secure funding for the system. CBD’s role in the project was facilitating the operational and technical implementation of the system and serving as a liaison between hospitals, health departments, and DataSphere, LLC. DataSphere, LLC owns and operates BioDefend and was responsible for the technical setup and maintenance of the system. The study addressed the feasibility of automated data collection by healthcare facilities and issues related to implementation of a syndromic surveillance system.

 

Objective

The purpose of this study was to evaluate the implementation of the BioDefend syndromic surveillance system for its use.

Submitted by elamb on
Description

Emerging infections, both natural and intentional, have provided an impetus for improved disease surveillance and response. The recognition of the interdependence of health care systems and public health infrastructure provides an opportunity to expand beyond traditional disease-based surveillance to a more comprehensive, integrated approach that leverages existing electronic information. The Veterans Affairs (VA) hospital system is uniquely positioned to perform multi-institutional enhanced electronic surveillance. A wealth of electronic information and technology resources are available in all VA hospitals and their associated clinics, as each facility uses the same standardized Computer Patient Record System. Influenza-like illness (ILI) is a common clinical syndrome of diverse etiology that presents with respiratory and systemic symptoms. The NC health department mandates the reporting of ILI from emergency departments to facilitate monitoring of seasonal ILI and serve as an important component of pandemic preparedness. Existing surveillance systems utilize an ICD-9 respiratory code screen and subsequent manual chart review which is timeconsuming and insensitive. Automated medical record review using more comprehensive electronic data may improve the system’s timeliness and efficiency.

 

Objective

To use data collected by NC-VET to create an automated ILI surveillance program and compare its accuracy and efficiency to the existing program.

Submitted by elamb on