About
My work has moved from numerical analysis and causal network inference toward computational biology, biostatistics, and clinical AI. The connecting question is how much confidence a model warrants for a specific scientific or clinical decision.
The arc
Foundations. I began in applied mathematics, studying numerical methods for nonlinear dynamical systems, solver design, convergence, and stability. My Ph.D. at West Virginia University extended that foundation to ensemble Bayesian inference of heterocellular networks in cancer.
Biomedical machine learning. At the WVU Cancer Institute, I worked on generative models, Bayesian networks, and deep learning for biological signal decomposition. An NSF fellowship in AI and machine learning for digital health supported the causal-network research that developed into BaMANI.
Clinical and translational work. At Duke, I moved into multiscale scientific machine learning, optimal-transport alignment, and NIH-supported congenital CMV modeling. My current work at MIT and Duke examines patient-level clinical AI reliability and epilepsy-surgery outcomes.
Appointments & education
Clinical research training
In 2026, I completed Duke Health CITI training in Human Subjects Research, Good Clinical Practice, HIPAA, and Clinical Data Privacy. I am also credentialed to use restricted PhysioNet datasets.
Research communities
I collaborate with the Duke Center for Human Systems Immunology and the Multiscale Immune Systems Modeling Center of Excellence. My professional affiliations include SITC, AACR, the Society for Mathematical Biology, and the American Mathematical Society.