Dynamic Insulin on Board: Incorporation of Circadian Insulin Sensitivity Variation

Abstract

Abbreviations: (BB) basal-bolus, (BG) blood glucose, (BW) body weight, (CL) closed loop, (CR) carbohydrate ratio, (CSII) continuous subcutaneous insulin infusion, (CVGA) control-variability grid analysis, (dIOB) dynamic insulin on board, (IOB) insulin on board, (MPC) model predictive control, (sIOB) static insulin on board, (T1DM) type 1 diabetes mellitus Keywords: artificial pancreas, circadian rhythm, continuous subcutaneous insulin infusion, diabetes, insulin on board, insulin sensitivity

Gaussian process-based model predictive control of blood glucose for patients with type 1 diabetes mellitus

“…In our engineering approach, we model the circadian IS of the glucose metabolism by making the corresponding linear parameter k IS time-dependent. The circadian rhythm of this parameter is built on and adapted from the periodic function obtained in [5] (see Fig. 2D).…”

“…The glucose metabolism is subject to multiple influential factors that change in time (e.g. the quasi-periodic appearance of meals [4], the diurnal insulin sensitivity (IS) changes [5], and the irregular pattern of exercises [6]), which require This work was partially funded by the German Academic National Foundation and by the United States National Institutes of Health under grants DP3DK104057 and UC4DK108483. 1 Philips Chair for Medical Information Technology, Helmholtz-Institute, RWTH Aachen University, 52074 Aachen, Germany.…”

“…An adaptive controller using Gain Scheduling based on the BG concentration was developed in [11], whereas in [12] an estimate of the IS determines which controller was activated. In [5], the authors included the circadian IS of their metabolic model into an input constraint of a Model Predictive Controller (MPC) to enhance the control performance. A review of adaptive controllers for BG regulation, including Minimum Variance Control, Self-Tuning-Regulators, Linear Quadratic Regulators and Generalized Predictive Control, can be found in [13].…”

Circadian Variability of Insulin Sensitivity: Physiological Input for In Silico Artificial Pancreas

“…Moreover, it has been recently shown in simulation that introducing a temporal variation on S I may improve AP glucose control. 17 The current report therefore represents a natural and critical extension of our work by incorporating this information into the S2013, allowing us to run multiple-meal scenarios, thus enabling a more robust design of AP algorithms. The aims of this study are thus (1) to use the data that we previously generated 16 to set up a model of S I pattern variability in the T1DM population, (2) to incorporate this variability model into the S2013, and (3) to reproduce the variability observed in the experimental data and thus validate the refined simulator for future use.…”

Dynamic Rule-Based Algorithm to Tune Insulin-on-Board Constraints for a Hybrid Artificial Pancreas System

“…Several approaches have been tested to overcome such issue [11][12][13][14], but while an ultrarapid insulin analogue is not available [15], postprandial control using subcutaneous route will continue to be a challenging situation for closed-loop systems. Diverse studies have included estimations of insulin concentration in the body to avoid excessive insulin stacking [16][17][18][19]; however, hypo-and hyperglycemia are still a hazard for AP systems, and novel approaches are still required.…”