Research
My group develops rigorous and practical statistical approaches for complex health data—especially when decisions must generalize, remain interpretable, and work fairly across populations.
Methods that transport evidence across studies and populations, balance complex covariates, and learn treatment policies under real-world constraints.
Interpretable decision rules for choosing treatments and doses while accounting for fairness, uncertainty, and multiple clinical outcomes.
Auditable language models and multimodal prediction systems that remain robust across time, care settings, and patient populations.
Scalable statistical methods for association, mediation, clustering, and meta-analysis in microbial and host-genetic studies.
Active funding
Supported work across causal inference, precision medicine, microbiome science, and equitable clinical decision-making.
Earlier support
2021—2025 · $600K
2019—2022 · $730K
Recognition
NIH Long COVID Computational Challenge · Honorable Mention
Pediatric COVID-19 Data Challenge · Task 1 Winner
EHR DREAM Challenge · Mortality Prediction Winner
JASA Theory & Methods · Discussion Paper
Margolin Dissertation Award
ENAR Distinguished Student Paper Award