An ongoing question
Research
How context could help artificial pancreas systems.
What changes when a system understands the state behind its observations?
CBA Lab, Georgia Tech · Advised by Dr. Thomas Plötz
The direction
State-aware processing in artificial pancreas loops
Exploring the benefits of state-aware processing in artificial pancreas loops.
I’m interested in what the state behind an observation can add to a system’s understanding.
Current stage
This is a research direction. It is not a published paper, a clinically validated system, or a proven improvement in treatment outcomes.
Related work
Diatrend explores glucose trajectory prediction from continuous glucose monitoring (CGM) data, using PyTorch, TabTransformer/GNN, and FastAPI.
DiatrendLet’s talk
I welcome conversations about the problem, the approach, and what to explore next.
Email meA working principle
Be clear about what is observed,
what is inferred,
and what remains uncertain.