OJPHI: Vol. 5
Journal Information
Journal ID (publisher-id): OJPHI
ISSN: 1947-2579
Publisher: University of Illinois at Chicago Library
Article Information
©2013 the author(s)
open-access: This is an Open Access article. Authors own copyright of their articles appearing in the Online Journal of Public Health Informatics. Readers may copy articles without permission of the copyright owner(s), as long as the author and OJPHI are acknowledged in the copy and the copy is used for educational, not-for-profit purposes.
Electronic publication date: Day: 4 Month: 4 Year: 2013
collection publication date: Year: 2013
Volume: 5E-location ID: e106
Publisher Id: ojphi-05-106

Enabling Syndromic Surveillance in Pakistan
Ross Maciejewski*1
Shehzad Afzal2
Adam J. Fairfield1
Arif Ghafoor2
David S. Ebert2
Naeem Ayyaz4
Maaz Ahmed3
1Computer Science, Arizona State University, Tempe, AZ, USA;
2Purdue University, West Lafayette, IN, USA;
3King Edward Medical University, Lahore, Pakistan;
4University of Engineering and Technology, Lahore, Pakistan
*Ross Maciejewski, E-mail: rmacieje@asu.edu

Abstract
Objective

This work presents our first steps in developing a Global Real-time Infectious Disease Surveillance System (GRIDDS) employing robust and novel infectious disease epidemiology models with real-time inference and pre/exercise planning capabilities for Lahore, Pakistan. The objective of this work is to address the infectious disease surveillance challenges (specific to developing countries such as Pakistan) and develop a collaborative capability for monitoring and managing outbreaks of natural or manmade infectious diseases in Pakistan.

Methods

Utilizing our partner hospitals in the Lahore, Punjab area, we have begun developing a theoretical model of patient hospital visits with respect to diseases and syndromes within Pakistan. Our first thrust has focused on the collection, categorization and cleansing of data based on expert knowledge from our partnering institutions in Pakistan. Data consists of a patient’s home address and chief complaint which is then categorized into syndromes. Home addresses are geocoded utilizing the Google API with a resultant 72% accuracy. Unknown geolocations are aggregated only at the hospital level. Using this cleaned data, we employ methods similar to our previous work [1] on syndromic surveillance for early disease detection. Currently, we have collected over 600,000 patient records over 1.5 years.

We employ the use of choropleth maps, isopleth maps utilizing kernel density estimation of patient addresses, traditional control chart methods such as exponentially weighted moving averages (EWMA), and a non-parametric time series analysis approach (seasonal trend decomposition using loess smoothing (STL) [2]) which requires only 90 days of historical data to be put into operation. The time series models are deployed as part of a real-time surveillance system in which temporal anomalies over regions can be analyzed and disease outbreaks reported.

Results

Figure 1 illustrates our visual analytics toolkit in operation. Here we see the location of our partner hospital in the Lahore region. The hospital coverage is in the most populous location of the city, providing data as a sentinel site for the overall health of the city. Currently, our system employs the use of interactive filters and linked isopleth or choropleth maps with time series analysis on mouse over.

Conclusions

Currently our research has focused on one partner location within the city of Lahore. Our ongoing work is focusing on the adoption of such a system to other regions of the country and the development of disease spread simulations (particularly Dengue Fever) utilizing baseline data collected by our partners. We plan to integrate these models into our visual analytics system for real-time planning and simulation.


Acknowledgments

This work is supported by the Defense Threat Reduction Agency Award Number HDTRA1-10-1-0083.


References
[1]. Maciejewski R, et al. “Forecasting Hotspot – A Predictive Analytics Approach,”IEEE Transactions on Visualization and Computer Graphics 17(4):440–453.2011;
[2]. Hafen RP, et al. “Syndromic surveillance: STL modeling, visualizing and monitoring disease counts,”BMC Medical Informatics and Decision Making. 2009

Figures
ojphi-05-106f1.tif
[Figure ID: f1-ojphi-05-106]
Figure 1 

Article Categories:
  • ISDS 2012 Conference Abstracts

Keywords: syndromic surveillance, visual analytics, pakistan.




Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org