Epidemiology PhD Student Explores AI to Empower Local Public Health

By Erin Frick
ALBANY, N.Y. (March 25, 2025) — With a background as an English professor, University at Albany’s Heather Duncan forged an unconventional path to undertaking her PhD in epidemiology at the College of Integrated Health Sciences. Having completed her Master of Public Health (MPH) at UAlbany in 2024, Duncan is now spearheading the development of artificial intelligence tools to support county-level health departments — a project spurred by her involvement in UAlbany's National Science Foundation Innovation Corps (I-Corps) program last month.
Duncan's journey into public health began while pursuing her English PhD. As part of a course, Duncan read Steven Johnson's book "The Ghost Map," which chronicles the 1854 London cholera outbreak and the pioneering work of epidemiologist John Snow. Captivated by the story, Duncan found herself drawn to the field of epidemiology, intrigued by the way diseases and medical innovations can shape society and culture. In her spare time, Duncan began devouring books on epidemiology and disease.
After graduating, Duncan moved to China, where she spent several years as an assistant professor at a liberal arts college. Through her developing epidemiological lens, she paid close attention to differences in public health policy, and their effects on day-to-day life. After returning to the U.S. in 2019, Duncan joined UAlbany’s Writing and Critical Inquiry Program as a lecturer yet still felt drawn to public health. In 2021, she made the bold decision to embark on a new path.
"In the middle of the pandemic, I decided to begin transitioning my career, and epidemiology was the natural fit," Duncan said. "It just so happened that the Master of Public Health program here at UAlbany was perfect for what I wanted to do."
While pursuing her MPH, Duncan began to envision a future where technology, and specifically AI, could be leveraged to empower local public health efforts. Her current research focus is infodemiology — the epidemiological approach that relies on communications data like social media and search engine trends rather than traditional disease surveillance. Duncan became interested in this area as she saw the potential for AI applications to help solve data analysis challenges facing public health workers, particularly at the county level.
Ideas Powering Innovation
In February, Duncan participated in UAlbany's NSF I-Corps program — an entrepreneurial training experience run by the Office of Economic Development, Entrepreneurship and Industry Partnerships that is designed to help scientists and engineers transform ideas rooted in research into high-impact inventions. Duncan, who co-owns M&D Science Consulting and Communications with fellow UAlbany alum and former public health teaching faculty member Patrick Murphy, saw in the program an opportunity to explore data analysis gaps in county health departments and gain skills to begin translating her growing interest into a usable tool.
Over the course of the 3.5-week program, Duncan conducted 20 in-depth interviews with county-level health department professionals across New York State, gaining valuable insights into the data processing obstacles they face on a daily basis.
"Much of the data that county health departments need to process is funneled to them from the state and needs to be disaggregated to use at the county level," Duncan said. "It doesn’t help that many of the state-level systems don’t easily sync up with platforms used by county health departments, each of which has its own unique set of methods for data analysis and display."
Duncan believes that AI-powered tools could help public health workers more efficiently collate, process and share important health information.
"With my background in humanities, upon starting my MPH, I had no formal training in the statistical and computational skills that are critical to pursuing an advanced degree in epidemiology," Duncan said. "I turned to ChatGPT to help bolster my skills in areas such as coding and Excel and found it to be extremely helpful. This sparked my idea for developing a custom AI chatbot designed to support county-level public health professionals who work in data management, analysis and visualization."
Duncan is now laying the groundwork to develop such a tool.
"Consider the local health department dashboards that we followed during the pandemic to get updates on COVID-19 cases and vaccination rates in our communities," Duncan said. "At the county level, much of that information had to be processed and uploaded manually. Tools like ChatGPT can make it possible to generate code or assemble an intricate formula in Excel to accomplish these sorts of tasks more efficiently, by providing customized training and automation tools. It's incredibly important that we put these sorts of powerful tools in the hands of people working at the local level and at the state level."
Duncan’s vision could also provide a stable, centralized repository of institutional knowledge to streamline training when staff turnover occurs.
"Staff turnover is a common issue in the county health departments that I spoke to," Duncan said. "Each time someone new comes in, they have to be trained in data handling workflows that are often tedious, and time consuming to teach and learn. A dedicated AI tool could act as a ‘tutor’ to aid this process."
Duncan is actively building a team of collaborators at UAlbany, tapping into faculty expertise in AI, data science and public health. She is also exploring new funding opportunities through the university and state-level programs to support the development of her innovative tools.
"Many industries are very excited about AI and are trying to figure out ways to use these tools to become more efficient," Duncan said. “Yet, there seems to be a disconnect between the people who produce AIs and the concerns that we have as epidemiologists or as community public health workers. My goal is to use my expertise in the humanities and passion for public health to help bridge between the worlds of technology and community epidemiology to make accessible AI tools that serve health professionals at multiple levels.”