Skip to content

Risk Factor Discovery in Population Health

All of Us: ASCVD Prediction

Our lab is evaluating whether Fitbit biosignals can flag atherosclerotic cardiovascular disease earlier and guide targeted CAC imaging. Using All of Us data, we harmonize wearable streams with de-identified EHR labels, derive features such as heart rate variability, circadian activity regularity, and adherence gaps, then train and calibrate supervised models to classify ASCVD and estimate risk. We benchmark model transportability and fairness and align results with our DBDP pipelines for scalable deployment. Students and collaborators gain experience in time-series feature engineering, clinical data linkage, and end-to-end ML evaluation while collaborating with clinicians and population health teams.

Investigating the impact of the built environment on physical activity

We study how the places people live and move through each day influence how physically active they are. We combine data from Fitbit, which tracks steps throughout the day, with information about the surrounding environment, such as access to parks, sidewalks, public transit, and neighborhood features. By looking at these data together, we identify common patterns in how people move in different environments. We then use these patterns to estimate how changes in urban design—such as adding green space or improving walkability—might increase daily activity and reduce long-term health risks. This work helps cities design healthier neighborhoods and find ways for all communities to realize the benefits of physical-activity-aware urban design.

Funded by: 1F31HL179990-01

Lab members: Hayoung Jeong

Sleep and Circadian Rhythms

We are quantifying cardiovascular responses to circadian disruption and comparing digital biomarkers against more conventional circadian metrics for evaluating the cardiovascular effects of circadian disruptions. Such digital biomarkers are meant to provide longitudinal, accessible and individualized metrics for evaluating the cardiovascular risk resulting from circadian disruptions, including events such as shift work and daylight saving time transitions.

Related Publications: Wang, Will Ke. “Sleep health and wearable technology: Algorithmic development towards field-based sleep monitoring. Available from Dissertations & Theses @ Duke University; ProQuest Dissertations & Theses Global (2024)