Thu Nguyen
University of Maryland, MD·Public Health & Epidemiology
Why do Black, Latina and other racial and ethnic minority mothers and their babies still face worse health outcomes, from postpartum depression to preterm birth, and what could close that gap? Two projects tackle the question from different directions.
- Fields
Public Health & Epidemiology
Computer Science
Medicine & Clinical Research
- Topics
maternal and child health
health disparities
chatbot
- Grants
- 2 known grants, through Jun 2027
What they work on
- Develop the Rosie chatbot by building a large health information library and improving its natural language processing engine so it reliably understands the many ways new mothers phrase questions.
- Test in a randomized controlled trial, with mostly virtual recruitment, whether using Rosie lowers postpartum depression risk, reduces emergency room visits and increases well-baby visit attendance.
- Release an open-source package that lets other groups build their own health information chatbots.
- Track changes in area-level sentiment across the United States using machine learning models applied to online and social media data, and identify the local and national events tied to those changes.
- Determine whether shifts in area-level sentiment lead to changes in preterm birth and low birth weight, and identify protective factors that buffer these adverse birth outcomes.
More about the research
One builds and tests a chatbot that gives new mothers trustworthy health answers at the moment they need them. The other uses online and social media data to ask whether the mood of a community affects birth outcomes. Both combine epidemiology, the study of health patterns in populations, with computer science and machine learning.
Why it matters
Racial and ethnic minority mothers and infants face higher rates of postpartum depression, missed well-child checkups, preterm birth, low birth weight and death. A scalable chatbot could give these families fast, verified health information without costly home visits and protect them from web misinformation. Measures of community sentiment and protective factors could help public health agencies design population-level policies to reduce these disparities.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Bhuwan Dhingra
Duke University, NC·Computer Science
Develop automated methods to identify emerging and existing vaccine-related concerns in social media and blogs, and categorize them with a new taxonomy of concerns.
Daniel Walker
Ohio State University, OH·Public Health & Epidemiology
Test how integrated data systems, CHW learning collaboratives and resilience skill building affect timely prenatal and postpartum care, depression and preterm birth, using a 2x2 factorial cluster randomized trial across Ohio's Pathways Community Hubs.
Daniel Erchick
Johns Hopkins University, MD·Public Health & Epidemiology
Determine whether an early-pregnancy oral health package with antiseptic rinse, exams, hygiene training and supplies lowers the rate of small vulnerable newborns in a cluster-randomized trial of 2,280 pregnancies in rural Nepal.
The school
University of Maryland: 49 labs on PI Pages, $1.5B a year on research, 602 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.