Data-Driven Personalization of Automated Insulin Delivery
For individuals with type 1 diabetes using automated insulin delivery systems, this method improves glycemic control by adapting to daily variability, though the gains are incremental.
The paper presents a real-time, data-driven personalization approach for automated insulin delivery that adapts controller parameters using daily glycemic data, achieving a 2%, 3%, and 4% increase in time-in-range after 4, 8, and 17 weeks, respectively, in silico experiments on a 100-adult cohort.
Automated insulin delivery (AID) systems are often tuned for the population and offer limited online adaptation to the inter- and intrapatient variability in insulin needs caused by meal patterns, physical activity, and fluctuations in insulin sensitivity. We present a real-time, data-driven personalization approach that adapts controller parameters using the subject's daily glycemic data. The adaptation is formulated as projected gradient descent on a daily risk metric, where the gradient estimation is designed to attenuate noise and metabolic variability. We use contraction theory to validate the optimization framework and convergence of the closed-loop system under adaptation. In silico experiments on the 100-adult cohort of the FDA-accepted UVA/Padova T1D simulator show that our method improves glycemic risk and increases time-in-range (TIR, 70-180\,mg/dL) by 2%, 3%, and 4% after 4, 8, and 17 weeks, respectively, under variability in meal timing, meal size, and insulin sensitivity.