PhD Defense by Yifan Liu

Event Details
  • Date/Time:
    • Wednesday March 28, 2018
      2:00 pm - 4:00 pm
  • Location: Groseclose 403
  • Phone:
  • URL:
  • Email:
  • Fee(s):
  • Extras:
No contact information submitted.

Summary Sentence: Public Health Informatics - Biosurveillance and Operations Strategies

Full Summary: No summary paragraph submitted.

Title: Public Health Informatics - Biosurveillance and Operations Strategies

Advisor: Dr. Eva K. Lee

Committee members:
Dr. David Goldsman
Dr. Yajun Mei
Dr. Andy Sun
Dr. Ernest E. Smith (Centers for Disease Control and Prevention)
Date and Time: Wednesday, March 28, 2:00 PM

Location: Groseclose 403

Public health emergencies continue to pose serious threats to human lives and well-being of the world population. The 2014 Ebola virus outbreak in West Africa and the 2016 Zika virus outbreak in South and North America underscores the challenges post by some of these highly infectious diseases even as healthcare has been revolutionized over the past century with medical science and technology. Infectious diseases are one of the most common yet most serious types of public health crisis due to several reasons: high infectivity, difficulties in prevention, labor-intensive intervention, diagnosis, and treatment. This is amplified by the mobility of human population enabled by transportation technologies. To react efficiently and effectively to waves of infectious disease outbreaks, decision-makers, and public health officers must quickly analyze the current situation and predict the potential trend of an outbreak, evaluate multiple countermeasures and strategies, and implement an intervention plan that optimizes utilization of the available (limited) resource while achieving the best containment results. Additionally, these decisions need to be made in real time as any delay in this process may result in severe outcomes, including more infections, economic loss, and even lives loss.

In this dissertation, we advance infectious diseases models with applications in medical countermeasure operations and biosurveillance. We first propose a general-purpose modeling framework for infectious disease outbreaks and discuss how it can be used to facilitate decision making to achieve best containment results. We demonstrate how this modeling technique can be computerized and used to support public health operations.

In the first chapter, we describe a general-purpose modeling framework for infectious disease. By expanding and abstracting the traditional compartmental models widely used for disease propagation dynamics, our modeling framework is highly generic and can be viewed as a meta-model for compartmental models. We focus on the models for contact-based diseases and vector-borne diseases and apply it to two real-life scenarios: 1) the Zika virus outbreaks in Brazil and Puerto Rico, and 2) the avian influenza outbreaks in Egypt and the United States. We discuss how this generic modeling framework can be tailored and parameterized to suit the properties of different diseases at different regions and investigate the impact of different intervention strategies.

In the second chapter, we equip the disease modeling framework with an optimization engine to determine the optimal resource allocation strategy during an outbreak. We investigate the impact of point-of-dispensing (PODs) for rapid medical countermeasures dispensing during an outbreak. We analyze its usage for a regional smallpox outbreak in the state of Georgia and investigate vaccination strategies for maximum health protection.

In the third chapter, we explain the design and functionalities of an enterprise software package, the RealOpt suite. Designed at Georgia Tech with collaboration with the Centers for Diseases Control and Prevention since 2005, RealOpt is a modularized system. This chapter focuses on two aspects of RealOpt: 1) an expanded module known as RealOpt-Contingency, and 2) the use of RealOpt for evaluating health registry. We first discuss the design structure of the general RealOpt framework and its ease-of-usage by public health personnel for emergency preparedness and actual responses. We then demonstrate usage in multiple aspects of public health emergency response, including facility location optimization, resource allocation optimization, inventory and transportation management, and cost-effectiveness analysis of strategies. Finally, using the RealOpt suite, we analyze and compare the performances of different electronic registration technologies used during mass dispensing events and propose a prototypical registration tool that yields the highest efficiency and usability.


Additional Information

In Campus Calendar

Graduate Studies

Invited Audience
Faculty/Staff, Public, Graduate students, Undergraduate students
Phd Defense
  • Created By: Tatianna Richardson
  • Workflow Status: Published
  • Created On: Mar 14, 2018 - 8:19am
  • Last Updated: Mar 14, 2018 - 8:19am