skip to main content

Title: Differential Flatness based Run-to-Run Control of Blood Glucose for People with Type 1 Diabetes
The objective of this paper is to develop an open loop insulin input profile over a span of 24 hours which makes the glucose trajectory of a Type 1 diabetic person track a target glucose trajectory. The Bergman minimal model is chosen to represent the glucose-insulin dynamics which is shown to be differentially flat. An optimal control problem is posed by parameterizing the differentially flat output of the Bergman model using Fourier series, to result in an input profile that can be repeatedly administered every day. The solution to the optimization problem is then shown to present acceptable performance in terms of tracking and adhering to imposed constraints.  more » « less
Award ID(s):
Author(s) / Creator(s):
Date Published:
Journal Name:
2019 American Control Conference (ACC)
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. The focus of this paper is on the characterization of the uncertainties in the evolving states of a diabetic model, to permit a study of the impact of the time interval between insulin bolusing and meal initiation on hypo- and hyperglycemic events. A polynomial chaos based approach is used to characterize the independent uncertainties in the initial condition and meal size. Galerkin projection of the resulting equations reduce the stochastic differential equations to a set of deterministic equations. This forms the framework to optimize for the post bolusing time to initiate the meal. Two cost functions are considered which correspond to the postprandial hypoand hyperglycemic excursions of the blood glucose. Numerical results from the minimal Bergman model suggest a 13 and 14 minute interval between bolusing and the initiation of the meal. 
    more » « less
  2. Under healthy conditions, the pancreas responds to a glucose challenge by releasing insulin. Insulin suppresses lipolysis in adipose tissue, thereby decreasing plasma glycerol concentration, and it regulates plasma glucose concentration through action in muscle and liver. Insulin resistance (IR) occurs when more insulin is required to achieve the same effects, and IR may be tissue-specific. IR emerges during puberty as a result of high concentrations of growth hormone and is worsened by youth-onset obesity. Adipose, liver, and muscle tissue exhibit distinct dose-dependent responses to insulin in multi-phase hyperinsulinemic-euglycemic (HE) clamps, but the HE clamp protocol does not address potential differences in the dynamics of tissue-specific insulin responses. Changes to the dynamics of insulin responses would alter glycemic control in response to a glucose challenge. To investigate the dynamics of insulin acting on adipose tissue, we developed a novel differential-equations based model that describes the coupled dynamics of glycerol concentrations and insulin action during an oral glucose tolerance test in female adolescents with obesity and IR. We compared these dynamics to the dynamics of insulin acting on muscle and liver as assessed with the oral minimal model applied to glucose and insulin data collected under the same protocol. We found that the action of insulin on glycerol peaks approximately 67 min earlier ( p < 0.001) and follows the dynamics of plasma insulin more closely compared to insulin action on glucose as assessed by the parameters representing the time constants for insulin action on glucose and glycerol ( p < 0.001). These findings suggest that the dynamics of insulin action show tissue-specific differences in our IR adolescent population, with adipose tissue responding to insulin more quickly compared to muscle and liver. Improved understanding of the tissue-specific dynamics of insulin action may provide novel insights into the progression of metabolic disease in patient populations with diverse metabolic phenotypes. 
    more » « less
  3. The focus of this paper is on the development of an open loop controller for type 1 diabetic patients which is robust to meal and initial condition uncertainties in the presence of hypoand hyperglycemic constraints. Bernstein polynomials are used to parametrize the evolving uncertain blood-glucose. The unique bounding properties of these polynomials are then used to enforce the desired glycemic constraints. A convex optimization problem is posed in the perturbation space of the model and is solved repeatedly to sequentially converge on a sub-optimal solution. The proposed approach is demonstrated on the classic Bergman model for Type 1 diabetic patients. 
    more » « less
  4. Abstract

    The management of diabetes in a manner offering autonomous insulin therapy responsive to glucose‐directed need, and moreover with a dosing schedule amenable to facile administration, remains an ongoing goal to improve the standard of care. While basal insulins with reduced dosing frequency, even once‐weekly administration, are on the horizon, there is still no approved therapy that offers glucose‐responsive insulin function. Herein, a nanoscale complex combining both electrostatic‐ and dynamic‐covalent interactions between a synthetic dendrimer carrier and an insulin analogue modified with a high‐affinity glucose‐binding motif yields an injectable insulin depot affording both glucose‐directed and long‐lasting insulin availability. Following a single injection, it is even possible to control blood glucose for at least one week in diabetic swine subjected to daily oral glucose challenges. Measurements of serum insulin concentration in response to challenge show increases in insulin corresponding to elevated blood glucose levels, an uncommon finding even in preclinical work on glucose‐responsive insulin. Accordingly, the subcutaneous nanocomplex that results from combining electrostatic‐ and dynamic‐covalent interactions between a modified insulin and a synthetic dendrimer carrier affords a glucose‐responsive insulin depot for week‐long control following a single routine injection.

    more » « less
  5. Neural networks present a useful framework for learning complex dynamics, and are increasingly being considered as components to closed loop predictive control algorithms. However, if they are to be utilized in such safety-critical advisory settings, they must be provably "conformant" to the governing scientific (biological, chemical, physical) laws which underlie the modeled process. Unfortunately, this is not easily guaranteed as neural network models are prone to learn patterns which are artifacts of the conditions under which the training data is collected, which may not necessarily conform to underlying physiological laws. In this work, we utilize a formal range-propagation based approach for checking whether neural network models for predicting future blood glucose levels of individuals with type-1 diabetes are monotonic in terms of their insulin inputs. These networks are increasingly part of closed loop predictive control algorithms for "artificial pancreas" devices which automate control of insulin delivery for individuals with type-1 diabetes. Our approach considers a key property that blood glucose levels must be monotonically decreasing with increasing insulin inputs to the model. Multiple representative neural network models for blood glucose prediction are trained and tested on real patient data, and conformance is tested through our verification approach. We observe that standard approaches to training networks result in models which violate the core relationship between insulin inputs and glucose levels, despite having high prediction accuracy. We propose an approach that can learn conformant models without much loss in accuracy. 
    more » « less