Understanding Bias Propagation in AI Model Predictions *Now Hiring a PhD Graduate Research Assistant*

March 2023-Present

Dr. Anderson and team are investigating how the artificial intelligence disease modeling pipeline can lead from biased data to biased predictions and to derive solutions that mitigate this bias. A case study will focus on infectious disease modeling. This US-Australian collaborative project is funded by The Commonwealth Scientific and Industrial Research Organization (CSIRO) and US National Science Foundation (NSF) based the potential to deliver transformative research towards responsible and equitable AI. The US-based PI is an Associate Professor of Computer Science at Emory University. This project is supported by the National Science Foundation (#2302970)

*Dr. Anderson is looking for a graduate research assistant to support this project. Interested applicants should contact her directly.

Predicting spillover of zoonotic diseases from urban wildlife to human populations.

August 2022-Present

Zoonotic diseases account for approximately 75% of all emerging infections and this is expected to rise. Yet the factors that trigger zoonotic disease emergence and spread remain poorly understood, especially in urban settings where wildlife and humans are frequently in contact. As part of an interdisciplinary collaboration with an epidemiologist at GMU and an urban ecologist at University of Maryland, Dr. Anderson and team are investigating the potential for zoonotic transmission of diseases between humans and urban wildlife and develop a modeling framework that can more broadly predict the potential of disease transmission. This research is supported by seed funding from GMU’s Office of Research Innovation and Economic Impact.

Data-Driven Modeling to Improve Understanding of Human Behavior, Mobility, and Disease Spread

May 2021-Present

Dr. Anderson and team are investigating how the inclusion of how more realistic representations of human behavior can improve spatial agent-based models (ABM) of disease spread. The project focuses on the development of a conceptual data-driven modeling framework that acknowledges the spatial and temporal heterogeneity of human behavior and tightly-couples behavior to mobility, human interaction, and disease dynamics. The modeling framework will be transferrable and scalable so that it can be used to simulate a variety of diseases in different study areas. This project is funded by the National Science Foundation (#2109647)

The Role of the Urban Built Environment in Breast Cancer Mortality Health Disparities

May-August 2022

Dr. Anderson and team are exploring the relationship between the urban built environment and breast cancer mortality health disparities in the United States. Health disparities are driven by complex and often interrelated factors. The built environment in which an individual is lives, including greenspace, health care providers, housing, walkability, healthy food choices, profoundly affects health and wellness. Despite recent commentaries that urge a better understanding of how

and why the environment matters to an individual’s health, the assessment of built environmental factors in relation to breast cancer survival is relatively unexplored. This project was funded by the Provost Summer Team Impact Project Award.