About

I am a Research Assistant Professor in the Department of Health Outcomes and Biomedical Informatics at the University of Florida, working in biomedical informatics and data science with real-world clinical data.

My work uses electronic health records and administrative claims to improve how clinical evidence is produced: how cohorts and eligibility are defined, how care and outcomes are measured, and how predictions and study plans are checked against what the data can actually observe. Current work centers on adolescent and maternal mental health, on whether AI can help researchers specify sound data requests, and on estimating trial feasibility from real-world data.

I came to informatics from biostatistics. Before joining the faculty I spent several years at UF building and maintaining research datasets, quality measures and models, including quality measurement for the Texas Medicaid and CHIP managed care report cards, machine learning models for early prediction of Alzheimer's disease and related dementias, and clinical NLP for de-identifying notes. My doctoral research studied how trial eligibility criteria affect who can take part in clinical trials.

Education

  1. PhD, Biomedical InformaticsUniversity of Florida
  2. MS, BiostatisticsUniversity of Florida2015
  3. BS, BiotechnologySun Yat-sen University, China2012

Positions

  1. Research Assistant ProfessorDepartment of Health Outcomes and Biomedical Informatics, University of FloridaCurrent
  2. Data Scientist IVDepartment of Health Outcomes and Biomedical Informatics, University of Florida2023 to 2025
  3. Data Management Analyst IIIDepartment of Health Outcomes and Biomedical Informatics, University of Florida2017 to 2023
  4. BiostatisticianPalmer Center for Chiropractic Research, Palmer College of Chiropractic2015 to 2017

Recognition

  1. Second place, Genevieve Melton-Meaux Student Paper CompetitionAMIA 2019 Annual Symposium, for a study of an a priori patient-trial generalizability score using real-world data2019

Methods and tools

Observational study design, cohort construction and computable phenotypes, survival and longitudinal models, machine learning and clinical NLP. Python, SAS, R and SQL; data visualization with Python, D3 and Tableau.

Contact

Email is best: [email protected]. I am glad to hear about collaborations in real-world data, mental health services research and clinical trial design.