In silico experiments of human eating patterns: Making a case for behavior-informed control of T1DM

Automatic control of Type 1 Diabetes Mellitus (T1DM) with subcutaneous (SC) measurement of glucose concentration and subcutaneous (SC) insulin infusion is of great interest within the diabetes technology research community. Realistic computer simulation can provide invaluable information about the safety and limitations of closed-loop control algorithms, can guide and focus the emphasis of clinical studies, and can out-rule ineffective control scenarios in a cost-effective manner prior to human use. In this work, we introduce a stochastic simulation model of patient eating behavior that can be employed in silico to test the ability of closed loop control strategies to adapt to individual physiologic characteristics and to the behavioral profile of each person. The design of this simulation model is motivated by the realization that historical data about a patient's eating behavior can be used to characterize the timing, likelihood, and magnitude of a meal disturbance within pre-specified meal regimes (e.g. breakfast). We demonstrate the use of our simulation model in evaluating a recent behaviorally-oriented T1DM control strategy introduced in Patek et al. (2009a).