Minh-Ha Nguyen

Minh-Ha Nguyen

About

My work is about one question: when can you trust a conclusion drawn from messy data?

In pharmacoepidemiology, that means asking whether an apparent drug risk is real or an artifact of how the study was designed. In AI, it means asking whether a benchmark score actually measures what we think it does. Different domains, same problem: when should you believe the number?

My doctoral work tackles this in Navy submariners exposed to chronic low-dose radiation. I study whether that exposure increases Parkinson's disease risk. The challenge is that the men with the highest exposure also served the longest and were often the healthiest to begin with, exactly the kind of selection that can hide a real hazard.

Background

I came to epidemiology through pharmacy. I hold a Master of Pharmacy (MPharm) and am pursuing doctoral studies in Epidemiology at Vanderbilt University, with training across clinical pharmacy, epidemiology, and biostatistics. Off the clock I read about deep learning, algorithm design, and reinforcement learning, partly for fun, mostly because I've always been obsessed with optimizing things in general.

Contact