About

I spent twelve years building software inside hospitals before I started a Ph.D., and that order matters.

I started as a junior developer on an electronic health record system in 2009, ended up leading that team, then started a health informatics group, and eventually ran a hospital IT department. So I saw what a hospital’s data actually looks like long before I saw it in a paper. It is incomplete. It was collected under pressure by people with other things to do. It is shaped by which patients happen to walk through that particular door.

In 2022 I went to Missouri S&T to work on the general version of that problem. My dissertation, “Adaptive Federated Learning for Survival Analysis in Smart Healthcare,” was defended in September 2026 in the CReWMaN lab under Sajal K. Das. It runs to ten chapters, and every one of them is some form of two questions. How do institutions that cannot pool their data still learn from each other, and how do you predict when an event will arrive rather than only whether it will. Hospitals are the case I worked it out on, and the question there has a face: whose experience should count when a doctor is predicting what happens to the person in front of them, and how much.

There is an awkward problem sitting inside federated learning, and it is not a medical problem. To know whether another institution’s contribution is helping your model, you would like to look at what they sent you. That is a reasonable thing to want. It is also the exact thing federated learning exists to avoid. Most of my work is about living with that tension rather than pretending it is not there. The timing half travels just as easily. Outside medicine the same models answer when a borrower defaults or when a customer stops renewing, and insurers and maintenance teams ask their own version of the question. I have not worked in those industries, and I am not going to pretend otherwise. What transfers is the method.

Alongside the research I keep shipping software, which I think keeps the research honest. I built a production oncology EMR for a practicing oncologist. I built a cardiovascular screening tool for community health workers in a cross-university field study, offline first, because the people using it are often somewhere with no connection.

I also went through NSF I-Corps in 2023, which mostly meant sitting with 157 clinicians, researchers and industry people and asking them what was actually in their way. That changed the work more than any experiment did.

I live in Rolla, Missouri.