PhD Defense | Efficient and Pragmatic Decisions Under Uncertainty in Healthcare

Title: Efficient and Pragmatic Decisions Under Uncertainty in Healthcare

Date: July 3, 2025 

Time: 10am

Paul Horton

Machine Learning PhD Student

H. Milton Stewart School of Industrial and Systems Engineering
Georgia Institute of Technology

Committee

1 Dr. David Goldsman (Advisor)

2 Dr. Christos Alexopoulos

3 Dr. Xiaoming Huo

4 Dr. Yajun Mei

Abstract

As personalized medicine impacts both the costs and benefits of healthcare, there is a growing need to make pragmatic decisions under resource constraints. In this proposal, I define a plan to research problems in the healthcare system which will address gaps in the literature related to efficient decisions. Among these, I will address a broad class of problems including the selection of a treatment from several alternatives, testing the performance of a diagnostic device with non-rectangular decision boundaries, and using imperfect references to quantify performance. A common theme among these problems is the formulation of a function to optimize, such as the relative benefit of a good decision, while constrained by regulations or penalized by resource consumption. Additionally, these problems include a temporal component as the time to make a decision has an associated cost along with multiple subpopulations to align with concepts in personalized medicine.