A critical assessment of algorithms and challenges in the development of a closed-loop artificial pancreas

Artificial pancreas: an emerging approach to treat Type 1 diabetes

Intensive insulin therapy aimed at achieving normal glucose levels significantly reduces the complications that are associated with diabetes but is also associated with an increased risk of low glucose levels (hypoglycemia). The growing use of continuous glucose monitors has stimulated the development of the artificial pancreas, a closed-loop insulin-delivery system aimed at restoring near-normal glucose levels while reducing the risk of hypoglycemia. The artificial pancreas comprises three components: a continuous glucose monitor, an insulin infusion pump and a control algorithm delivering insulin according to real-time glucose readings. In this article, we review closed-loop glucose control, including its components, development, testing and clinical application.

Can we really close the loop and how soon? Accelerating the availability of an artificial pancreas: a roadmap to better diabetes outcomes

Development of a closed-loop artificial pancreas has been a long-time goal that could transform diabetes management. The primary limitation until recent years was the lack of a robust and portable continuous glucose sensor. There has been significant progress over the past 5 years in the development and commercialization of continuous glucose monitoring (CGM) devices. Used adjunctively, CGM has been demonstrated to add significant value in improving diabetes management by increasing time spent in glycemic targets and improving overall glycemic control. However, these benefits are limited by the human user's finite capacity to respond to the data provided by the device. By automating even a portion of the insulin delivery functionality of combined sensor/pump systems via computer algorithm, impending excursions could be handled more quickly and effectively. This review will describe very promising preliminary closed-loop studies, describe a potential roadmap to an artificial pancreas that will be safe and effective, and propose a solution-a hypo- and hyperglycemia minimizing control-to-range approach-that may allow for near-term delivery of a semiautomated system to people with diabetes.

Safety constraints in an artificial pancreatic beta cell: an implementation of model predictive control with insulin on board

BACKGROUND

Type 1 diabetes mellitus (T1DM) is characterized by the destruction of pancreatic beta cells, resulting in the inability to produce sufficient insulin to maintain normoglycemia. As a result, people with T1DM depend on exogenous insulin that is given either by multiple daily injections or by an insulin pump to control their blood glucose. A challenging task is to design the next step in T1DM therapy: a fully automated insulin delivery system consisting of an artificial pancreatic beta cell that shall provide both safe and effective therapy. The core of such a system is a control algorithm that calculates the insulin dose based on automated glucose measurements.

METHODS

A model predictive control (MPC) algorithm was designed to control glycemia by controlling exogenous insulin delivery. The MPC algorithm contained a dynamic safety constraint, insulin on board (IOB), which incorporated the clinical values of correction factor and insulin-to-carbohydrate ratio along with estimated insulin action decay curves as part of the optimal control solution.

RESULTS

The results emphasized the ability of the IOB constraint to significantly improve the glucose/insulin control trajectories in the presence of aggressive control actions. The simulation results indicated that 50% of the simulations conducted without the IOB constraint resulted in hypoglycemic events, compared to 10% of the simulations that included the IOB constraint.

CONCLUSIONS

Achieving both efficacy and safety in an artificial pancreatic beta cell calls for an IOB safety constraint that is able to override aggressive control moves (large insulin doses), thereby minimizing the risk of hypoglycemia.

Is an automatic pump suspension feature safe for children with type 1 diabetes? An exploratory analysis with a closed-loop system

OBJECTIVES

It has been proposed that the first step towards a closed-loop artificial pancreas might be to use a continuous glucose sensor to automatically suspend the basal insulin delivery based on projected low sensor glucose values.

METHODS

We reviewed our recent experience with an artificial pancreas system, utilizing a proportional-integrative-derivative (PID) algorithm, in 17 adolescents with type 1 diabetes (T1D) to assess the safety and efficacy of this maneuver.

RESULTS

During 34 h of closed-loop automated insulin delivery, 18 pump suspensions > or =60 min (90 +/- 18 min) occurred in eight subjects. Sensor glucose levels fell from 159 +/- 42 mg/dL to a nadir of 72 +/- 13 mg/dL. Corresponding plasma glucose levels fell from 168 +/- 51 to 72 +/- 16 mg/dL, with values <60 mg/dL recorded in only four of the 18 events.

CONCLUSIONS

These data suggest that automatic pump suspension using the PID algorithm may be an effective means to prevent hypoglycemia in youth with T1D.

Effectiveness of Intravenous Infusion Algorithms for Glucose Control in Diabetic Patients Using Different Simulation Models

The effectiveness of closed-loop insulin infusion algorithms is assessed for three different mathematical models describing insulin and glucose dynamics within a Type I diabetes patient. Simulations are performed to assess the effectiveness of proportional plus integral plus derivative (PID) control, feedforward control, and a physiologically-based control system with respect to maintaining normal glucose levels during a meal and during exercise. Control effectiveness is assessed by comparing the simulated response to a simulation of a healthy patient during both a meal and exercise and establishing maximum and minimum glucose levels and insulin infusion levels, as well as maximum duration of hyperglycemia. Controller effectiveness is assessed within the minimal model, the Sorensen model, and the Hovorka model. Results showed that no type of control was able to maintain normal conditions when simulations were performed using the minimal model. For both the Sorensen model and the Hovorka model, proportional control was sufficient to maintain normal glucose levels. Given published clinical data showing the ineffectiveness of PID control in patients, the work demonstrates that controller success based on simulation results can be misleading, and that future work should focus on addressing the model discrepancies.

In silico preclinical trials: methodology and engineering guide to closed-loop control in type 1 diabetes mellitus

This article sets forth guidelines for in silico (simulation-based) proof-of-concept testing of artificial pancreas control algorithms. The goal was to design a test procedure that can facilitate regulatory approval [e.g., Food and Drug Administration Investigational Device Exemption] for General Clinical Research Center experiments without preliminary testing on animals. The methodology is designed around a software package, based on a recent meal simulation model of the glucose-insulin system. Putting a premium on generality, this document starts by specifying a generic, rather abstract, meta-algorithm for control. The meta-algorithm has two main components: (1) patient assessment and tuning of control parameters, i.e., algorithmic processes for collection and processing patient data prior to closed-loop operation, and (2) controller warm-up and run-time operation, i.e., algorithmic processes for initializing controller states and managing blood glucose. The simulation-based testing methodology is designed to reveal the conceptual/mathematical operation of both main components, as applied to a large population of in silico patients with type 1 diabetes mellitus.

Architecture of a wireless Personal Assistant for telemedical diabetes care

PURPOSE

Advanced information technologies joined to the increasing use of continuous medical devices for monitoring and treatment, have made possible the definition of a new telemedical diabetes care scenario based on a hand-held Personal Assistant (PA). This paper describes the architecture, functionality and implementation of the PA, which communicates different medical devices in a personal wireless network.

DESCRIPTION OF THE SYSTEM

The PA is a mobile system for patients with diabetes connected to a telemedical center. The software design follows a modular approach to make the integration of medical devices or new functionalities independent from the rest of its components. Physicians can remotely control medical devices from the telemedicine server through the integration of the Common Object Request Broker Architecture (CORBA) and mobile GPRS communications. Data about PA modules' usage and patients' behavior evaluation come from a pervasive tracing system implemented into the PA.

RESULTS AND DISCUSSION

The PA architecture has been technically validated with commercially available medical devices during a clinical experiment for ambulatory monitoring and expert feedback through telemedicine. The clinical experiment has allowed defining patients' patterns of usage and preferred scenarios and it has proved the Personal Assistant's feasibility. The patients showed high acceptability and interest in the system as recorded in the usability and utility questionnaires. Future work will be devoted to the validation of the system with automatic control strategies from the telemedical center as well as with closed-loop control algorithms.

Diabetes: Models, Signals, and Control

The control of diabetes is an interdisciplinary endeavor, which includes a significant biomedical engineering component, with traditions of success beginning in the early 1960s. It began with modeling of the insulin-glucose system, and progressed to large-scale in silico experiments, and automated closed-loop control (artificial pancreas). Here, we follow these engineering efforts through the last, almost 50 years. We begin with the now classic minimal modeling approach and discuss a number of subsequent models, which have recently resulted in the first in silico simulation model accepted as substitute to animal trials in the quest for optimal diabetes control. We then review metabolic monitoring, with a particular emphasis on the new continuous glucose sensors, on the analyses of their time-series signals, and on the opportunities that they present for automation of diabetes control. Finally, we review control strategies that have been successfully employed in vivo or in silico, presenting a promise for the development of a future artificial pancreas and, in particular, discuss a modular architecture for building closed-loop control systems, including insulin delivery and patient safety supervision layers. We conclude with a brief discussion of the unique interactions between human physiology, behavioral events, engineering modeling and control relevant to diabetes.

Treatment of diabetes mellitus using an external insulin pump: the state of the art

The aim of diabetes treatment is to achieve tight glucose control to avoid the development of chronic diabetic complications while reducing the frequency of hypoglycaemic episodes. Continuous subcutaneous insulin infusion (CSII) using an external pump is an intensive diabetes therapy recognized to improve metabolic control and glycaemic instability, and to reduce the frequency of severe hypoglycaemia. For years, the theoretical advantages of the insulin pump (constancy of basal delivery, adjustable basal rates, and low insulin depots allowing the reduction of glycaemic variability) have contributed to its reported superiority compared with multiple daily injections (MDI). However, insulin pump therapy is now challenged by new MDI regimens based on long-acting insulin analogues that could replace the use of CSII. As a consequence, health professionals now have to determine which patients are likely to benefit the most from CSII. Recently, several studies reported that children and adolescents, and patients whose blood glucose imbalance was initially the most pronounced with basal-bolus regimens, would particularly benefit from CSII. Other indications were also proposed in marginal clinical situations with highly selected patients in whom a significant improvement of blood glucose was demonstrated. Finally, the use of CSII in type 2 diabetic patients now appears to be a good alternative to the ineffective MDI regimens observed in some of these patients. However, past experience with CSII indicates that candidates for insulin pump therapy must be carefully selected and strongly motivated to improve their glucose control. Use of CSII also requires strict medical supervision by physicians and a regular programme of patient education by paramedical teams, to ensure optimal responsible use of this technique by healthcare professionals.

The INCA system: a further step towards a telemedical artificial pancreas

Biomedical engineering research efforts have accomplished another level of a "technological solution" for diabetes: an artificial pancreas to be used by patients and supervised by healthcare professionals at any time and place. Reliability of continuous glucose monitoring, availability of real-time programmable insulin pumps, and validation of safe and efficient control algorithms are critical components for achieving that goal. Nevertheless, the development and integration of these new technologies within a telemedicine system can be the basis of a future artificial pancreas. This paper introduces the concept, design, and evaluation of the "intelligent control assistant for diabetes, INCA" system. INCA is a personal digital assistant (PDA)-based personal smart assistant to provide patients with closed-loop control strategies (personal and remote loop), based on a real-time continuous glucose sensor (Guardian RT, Medtronic), an insulin pump (D-TRON, Disetronic Medical Systems), and a mobile general packet radio service (GPRS)-based telemedicine communication system. Patient therapeutic decision making is supervised by doctors through a multiaccess telemedicine central server that provides to diabetics and doctors a Web-based access to continuous glucose monitoring and insulin infusion data. The INCA system has been technically and clinically evaluated in two randomized and crossover clinical trials showing an improvement on glycaemic control of diabetic patients.

Closed-loop control and advisory mode evaluation of an artificial pancreatic Beta cell: use of proportional-integral-derivative equivalent model-based controllers

BACKGROUND

Using currently available technology, it is possible to apply modern control theory to produce a closed-loop artificial beta cell. Novel use of established control techniques would improve glycemic control, thereby reducing the complications of diabetes. Two popular controller structures, proportional-integral-derivative (PID) and model predictive control (MPC), are compared first in a theoretical sense and then in two applications.

METHODS

The Bergman model is transformed for use in a PID equivalent model-based controller. The internal model control (IMC) structure, which makes explicit use of the model, is compared with the PID controller structure in the transfer function domain. An MPC controller is then developed as an optimization problem with restrictions on its tuning parameters and is shown to be equivalent to an IMC controller. The controllers are tuned for equivalent performance and evaluated in a simulation study as a closed-loop controller and in an advisory mode scenario on retrospective clinical data.

RESULTS

Theoretical development shows conditions under which PID and MPC controllers produce equivalent output via IMC. The simulation study showed that the single tuning parameter for the equivalent controllers relates directly to the closed-loop speed of response and robustness, an important result considering system uncertainty. The risk metric allowed easy identification of instances of inadequate control. Results of the advisory mode simulation showed that suitable tuning produces consistently appropriate delivery recommendations.

CONCLUSION

The conditions under which PID and MPC are equivalent have been derived. The MPC framework is more suitable given the extensions necessary for a fully closed-loop artificial beta cell, such as consideration of controller constraints. Formulation of the control problem in risk space is attractive, as it explicitly addresses the asymmetry of the problem; this is done easily with MPC.

Evaluating the efficacy of closed-loop glucose regulation via control-variability grid analysis

BACKGROUND

Advancements in subcutaneous continuous glucose monitoring and subcutaneous insulin delivery are stimulating the development of a minimally invasive artificial pancreas that facilitates optimal glycemic regulation in diabetes. The key component of such a system is the blood glucose controller for which different design strategies have been investigated in the literature. In order to evaluate and compare the efficacy of the various algorithms, several performance indices have been proposed.

METHODS

A new tool-control-variability grid analysis (CVGA)-for measuring the quality of closed-loop glucose control on a group of subjects is introduced. It is a method for visualization of the extreme glucose excursions caused by a control algorithm in a group of subjects, with each subject presented by one data point for any given observation period. A numeric assessment of the overall level of glucose regulation in the population is given by the summary outcome of the CVGA.

RESULTS

It has been shown that CVGA has multiple uses: comparison of different patients over a given time period, of the same patient over different time periods, of different control laws, and of different tuning of the same controller on the same population.

CONCLUSIONS

Control-variability grid analysis provides a summary of the quality of glycemic regulation for a population of subjects and is complementary to measures such as area under the curve or low/high blood glucose indices, which characterize a single glucose trajectory for a single subject.

The future of open- and closed-loop insulin delivery systems

We have analysed several aspects of insulin-dependent diabetes mellitus, including the glucose metabolic system, diabetes complications, and previous and ongoing research aimed at controlling glucose in diabetic patients. An expert review of various models and control algorithms developed for the glucose homeostasis system is presented, along with an analysis of research towards the development of a polymeric insulin infusion system. Recommendations for future directions in creating a true closed-loop glucose control system are presented, including the development of multivariable models and control systems to more accurately describe and control the multi-metabolite, multi-hormonal system, as well as in-vivo assessments of implicit closed-loop control systems.

Development of a clinical type 1 diabetes metabolic system model and in silico simulation tool

OBJECTIVES

The goal of this study was to develop a system model of type 1 diabetes for the purpose of in silico simulation for the prediction of long-term glycemic control outcomes.

METHODS

The system model was created and identified on a physiological cohort of virtual type 1 diabetes patients (n = 40). Integral-based identification was used to develop (n = 40) insulin sensitivity profiles.

RESULTS

The n = 40 insulin sensitivity profiles provide a driving input for virtual patient trials using the models developed. The identified models have a median (90% range) absolute percentage error of 1.33% (0.08-7.20%). The median (90% range) absolute error was 0.12 mmol/liter (0.01-0.56 mmol/liter). The model and integral-based identification of SI captured all patient dynamics with low error, which would lead to more physiological behavior simulation.

CONCLUSIONS

A simulation tool incorporating n = 40 virtual patient data sets to predict long-term glycemic control outcomes from clinical interventions was developed based on a physiological type 1 diabetes metabolic system model. The overall goal is to utilize this model and insulin sensitivity profiles to develop and optimize self-monitoring blood glucose and multiple daily injection therapy.

In silico simulation of long-term type 1 diabetes glycemic control treatment outcomes

OBJECTIVES

The goals of this study were to develop (1) a safe and effective protocol for the clinical control of type 1 diabetes using conventional self-monitoring blood glucose (SMBG) measurements and multiple daily injections with insulin analogues, and (2) an in silico simulation tool of type 1 diabetes to predict long-term glycemic control outcomes of clinical interventions.

METHODS

The virtual patient method was used to develop a simulation tool for type 1 diabetes using data from a type 1 diabetes patient cohort (n = 40). The tool was used to test the adaptive protocol (AC) and a conventional intensive insulin therapy (CC) against results from a representative control cohort. Optimal and suboptimal basal insulin replacements were evaluated as a function of SMBG frequency in conjunction with the (AC and CC) prandial control protocols.

RESULTS

In long-term glycemic control, the AC protocol significantly decreased hemoglobin A1c in conditions of suboptimal basal insulin replacement for SMBG frequencies > or = 6/day, and reduced the occurrence of mild and severe hypoglycemia by 86-100% over controls, over all SMBG frequencies in conditions of optimal basal insulin.

CONCLUSIONS

A simulation tool to predict long-term glycemic control outcomes from clinical interventions has been developed to test a novel, adaptive control protocol for type 1 diabetes. The protocol is effective and safe compared to conventional intensive insulin therapy and controls. As fear of hypoglycemia is a large psychological barrier to glycemic control, the AC protocol may represent the next evolution of intensive insulin therapy to deliver increased glycemic control with increased safety. Further clinical or experimental validation is needed to fully prove the concept.

Fully automated closed-loop insulin delivery versus semiautomated hybrid control in pediatric patients with type 1 diabetes using an artificial pancreas

OBJECTIVE

The most promising beta-cell replacement therapy for children with type 1 diabetes is a closed-loop artificial pancreas incorporating continuous glucose sensors and insulin pumps. The Medtronic MiniMed external physiological insulin delivery (ePID) system combines an external pump and sensor with a variable insulin infusion rate algorithm designed to emulate the physiological characteristics of the beta-cell. However, delays in insulin absorption associated with the subcutaneous route of delivery inevitably lead to large postprandial glucose excursions.

RESEARCH DESIGN AND METHODS

We studied the feasibility of the Medtronic ePID system in youth with type 1 diabetes and hypothesized that small manual premeal "priming" boluses would reduce postprandial excursions during closed-loop control. Seventeen adolescents (aged 15.9 +/- 1.6 years; A1C 7.1 +/- 0.8%) underwent 34 h of closed-loop control; 8 with full closed-loop (FCL) control and 9 with hybrid closed-loop (HCL) control (premeal priming bolus).

RESULTS

Mean glucose levels were 135 +/- 45 mg/dl in the HCL group versus 141 +/- 55 mg/dl in the FCL group (P = 0.09); daytime glucose levels averaged 149 +/- 47 mg/dl in the HCL group versus 159 +/- 59 mg/dl in the FCL group (P = 0.03). Peak postprandial glucose levels averaged 194 +/- 47 mg/dl in the HCL group versus 226 +/- 51 mg/dl in the FCL group (P = 0.04). Nighttime control was similar in both groups (111 +/- 27 vs. 112 +/- 28 mg/dl).

CONCLUSIONS

Closed-loop glucose control using an external sensor and insulin pump provides a means to achieve near-normal glucose concentrations in youth with type 1 diabetes during the overnight period. The addition of small manual priming bolus doses of insulin, given 15 min before meals, improves postprandial glycemic excursions.

An improved PID switching control strategy for type 1 diabetes

In order for an "artificial pancreas" to become a reality for ambulatory use, a practical closed-loop control strategy must be developed and validated. In this paper, an improved PID control strategy for blood glucose control is proposed and critically evaluated in silico using a physiologic model of Hovorka et al. [1]. The key features of the proposed control strategy are: 1) a switching strategy for initiating PID control after a meal and insulin bolus; 2) a novel time-varying setpoint trajectory; 3) noise and derivative filters to reduce sensitivity to sensor noise; and 4) a practical controller tuning strategy. Simulation results demonstrate that proposed control strategy compares favorably to alternatives for realistic conditions that include meal challenges, incorrect carbohydrate meal estimates, changes in insulin sensitivity, and measurement noise.

An improved PID switching control strategy for type 1 diabetes

In order for an "artificial pancreas" to become a reality for ambulatory use, a practical closed-loop control strategy must be developed and critically evaluated. In this paper, an improved PID control strategy for blood glucose control is proposed and evaluated in silico using a physiologic model of Hovorka et al. The key features of the proposed control strategy are: (i) a switching strategy for initiating PID control after a meal and insulin bolus; (ii) a novel time-varying setpoint trajectory, (iii) noise and derivative filters to reduce sensitivity to sensor noise, and (iv) a systematic controller tuning strategy. Simulation results demonstrate that the proposed control strategy compares favorably to alternatives for realistic conditions that include meal challenges, incorrect carbohydrate meal estimates, changes in insulin sensitivity, and measurement noise.

Robust parameter estimation in a model for glucose kinetics in type 1 diabetes subjects

There is a significant push to develop closed-loop control systems to deliver insulin for type 1 diabetic subjects. As part of this process, mathematical models are required to test and validate the proposed algorithms. There are several published physiology-based models of glucose and insulin dynamics in the literature, however, all of them were derived using data from subjects without diabetes. For this particular study we have selected one of the recently published models, by Hovorka et al., replacing the subcutaneous insulin infusion model with the one described by Wilinska et al., Five subjects with type 1 diabetes underwent a hyperinsulinemic-euglycemic clamp with a meal challenge and corresponding subcutaneous insulin bolus. The data collected were used to fit the model parameters using global optimization methods. Our results show that the model is capable of describing the observed dynamics for type 1 subjects under the experimental conditions, and as such can be used to simulate subject behavior under the experimental conditions.

Glucose monitoring in acute care: technologies on the horizon

Current glucose monitoring technology appears inadequate for the management of diabetic surgical and in critically ill patients requiring intensive insulin therapy. Subcutaneous sensors measure interstitial fluid glucose, and this technology has not yet been shown to provide the timely and accurate measurements necessary for intravenous insulin administration in surgical and critical care patients on intensive insulin therapy. Technologies under development that may be more suitable for surgical and intensive care unit patients are the automated intermittent type glucose monitors and central catheter glucose monitors. Improved accuracy, patient safety, incorporation of control algorithms, and alleviation of added nursing labor are important factors for consideration with future acute care glucose monitors. Hospital costs for these monitors are difficult to estimate but may be relatively low if their use can be related to better patient outcome, reduced labor costs, and increased job satisfaction for the nursing staff.

A Feedforward-Feedback Glucose Control Strategy for Type 1 Diabetes Mellitus

As the "artificial pancreas" becomes closer to reality, automated insulin delivery based on real-time glucose measurements becomes feasible for people with diabetes. This paper is concerned with the development of novel feedforward-feedback control strategies for real-time glucose control and type 1 diabetes. Improved post-meal responses can be achieved by a pre-prandial snack or bolus, or by reducing the glucose setpoint prior to the meal. Several feedforward-feedback control strategies provide attractive alternatives to the standard meal insulin bolus and are evaluated in simulations using a physiological model.

Detection of a meal using continuous glucose monitoring: implications for an artificial beta-cell

OBJECTIVE

The purpose of this study was to introduce a novel meal detection algorithm (MDA) to be used as part of an artificial beta-cell that uses a continuous glucose monitor (CGM).

RESEARCH DESIGN AND METHODS

We developed our MDA on a dataset of 26 meal events using records from 19 children aged 1-6 years who used the MiniMed CGMS Gold. We then applied this algorithm to CGM records from a DirecNet pilot study of the FreeStyle Navigator continuous glucose sensor. During a research center admission, breakfast insulin was withheld for 1 h, and discrete glucose levels were obtained every 10 min after the meal.

RESULTS

Based on the Navigator readings, the MDA detected a meal at a mean time of 30 min from the onset of eating, at which time the mean serum glucose was 21 mg/dl above baseline (range 2-36 mg/dl), and >90% of meals were detected before the glucose had risen 40 mg/dl from baseline.

CONCLUSIONS

The MDA will enable automated insulin dosing in response to meals, facilitating the development of an artificial pancreas.

Linear quadratic gaussian-based closed-loop control of type 1 diabetes

We investigated the applicability of linear quadratic Gaussian (LQG) methodology to the subcutaneous blood glucose regulation problem. We designed an LQG-based feedback control algorithm using linearization of a previously published metabolic model of type 1 diabetes. A key feature of the controller is a Kalman filter used to estimate metabolic states of the patient based on continuous glucose monitoring. Insulin infusion is computed from linear quadratic regulator feedback gains applied to these estimates, generally seeking to minimize squared deviations from a target glucose concentration and basal insulin rate. We evaluated in silico subject-specific LQG control and compared it to preexisting proportional-integral-derivative control.

Analysis of algorithms for intensive care unit blood glucose control

Intensive care unit (ICU) blood glucose control algorithms were reviewed and analyzed in the context of linear systems theory and classical feedback control algorithms. Closed-loop performance was illustrated by applying the algorithms in simulation studies using an in silico model of an ICU patient. Steady-state and dynamic input-output analysis was used to provide insight about controller design and potential closed-loop performance. The proportional-integral-derivative, columnar insulin dosing (CID, Glucommander-like), and glucose regulation for intensive care patients (GRIP) algorithms were shown to have similar features and performance. The CID strategy is a time-varying proportional-only controller (no integral action), whereas the GRIP algorithm is a nonlinear controller with integral action. A minor modification to the GRIP algorithm was suggested to improve the closed-loop performance. Recommendations were made to guide control theorists on important ICU control topics worthy of further study.

Model predictive control of type 1 diabetes: an in silico trial

BACKGROUND

The development of artificial pancreas has received a new impulse from recent technological advancements in subcutaneous continuous glucose monitoring and subcutaneous insulin pump delivery systems. However, the availability of innovative sensors and actuators, although essential, does not guarantee optimal glycemic regulation. Closed-loop control of blood glucose levels still poses technological challenges to the automatic control expert, most notable of which are the inevitable time delays between glucose sensing and insulin actuation.

METHODS

A new in silico model is exploited for both design and validation of a linear model predictive control (MPC) glucose control system. The starting point is a recently developed meal glucose-insulin model in health, which is modified to describe the metabolic dynamics of a person with type 1 diabetes mellitus. The population distribution of the model parameters originally obtained in healthy 204 patients is modified to describe diabetic patients. Individual models of virtual patients are extracted from this distribution. A discrete-time MPC is designed for all the virtual patients from a unique input-output-linearized approximation of the full model based on the average population values of the parameters. The in silico trial simulates 4 consecutive days, during which the patient receives breakfast, lunch, and dinner each day.

RESULTS

Provided that the regulator undergoes some individual tuning, satisfactory results are obtained even if the control design relies solely on the average patient model. Only the weight on the glucose concentration error needs to be tuned in a quite straightforward and intuitive way. The ability of the MPC to take advantage of meal announcement information is demonstrated. Imperfect knowledge of the amount of ingested glucose causes only marginal deterioration of performance. In general, MPC results in better regulation than proportional integral derivative, limiting significantly the oscillation of glucose levels.

CONCLUSIONS

The proposed in silico trial shows the potential of MPC for artificial pancreas design. The main features are a capability to consider meal announcement information, delay compensation, and simplicity of tuning and implementation.

Practical issues in the identification of empirical models from simulated type 1 diabetes data

BACKGROUND

A model-based controller for an artificial beta-cell automatically regulates blood glucose levels based on available glucose measurements, insulin infusion and meal information, and model predictions of future glucose trends. Thus, the identification of simple, accurate models plays an important role in the development of an artificial beta-cell.

METHODS

Glucose data simulated from a nonlinear physiological model of type 1 diabetes are used to identify linear dynamic models of two types: autoregressive exogenous input (ARX) and output-error (OE) models. The model inputs are meal carbohydrates and exogenous insulin, which in practice are often administered simultaneously and in the same ratio, i.e., the insulin-to-carbohydrate ratio. The effect of modeling these inputs as impulses versus time-smoothed profiles ("transformed inputs") is explored in depth. The models are evaluated based on their ability to describe the data from which they were identified (i.e., calibration data) as well as independent data (i.e., validation data).

RESULTS

In general, the best models described their calibration data more accurately using transformed inputs (R(Cal) (2) = 71% for the ARX models and R (Cal) (2) = 78% for the OE models) than using impulse inputs (R (Cal) (2) = 14% for the ARX models and R (Cal) (2) = 70% for the OE models). The only model/input combination that resulted in consistently accurate validation fits was the ARX models using transformed inputs (39% <or= R (Val) (2) <or= 58%).

CONCLUSIONS

When identifying non-physiologically based models from diabetes data with simultaneous and proportional meals and insulin boluses, model accuracy is improved by modeling the inputs as time-smoothed profiles. Also, while OE models describe their calibration data very well, ARX models more accurately describe validation data. Their versatility makes ARX models a more attractive choice for implementation in a model-based controller of an artificial beta-cell.

Predictive monitoring for improved management of glucose levels

BACKGROUND

Recent developments and expected near-future improvements in continuous glucose monitoring (CGM) devices provide opportunities to couple them with mathematical forecasting models to produce predictive monitoring systems for early, proactive glycemia management of diabetes mellitus patients before glucose levels drift to undesirable levels. This article assesses the feasibility of data-driven models to serve as the forecasting engine of predictive monitoring systems.

METHODS

We investigated the capabilities of data-driven autoregressive (AR) models to (1) capture the correlations in glucose time-series data, (2) make accurate predictions as a function of prediction horizon, and (3) be made portable from individual to individual without any need for model tuning. The investigation is performed by employing CGM data from nine type 1 diabetic subjects collected over a continuous 5-day period.

RESULTS

With CGM data serving as the gold standard, AR model-based predictions of glucose levels assessed over nine subjects with Clarke error grid analysis indicated that, for a 30-minute prediction horizon, individually tuned models yield 97.6 to 100.0% of data in the clinically acceptable zones A and B, whereas cross-subject, portable models yield 95.8 to 99.7% of data in zones A and B.

CONCLUSIONS

This study shows that, for a 30-minute prediction horizon, data-driven AR models provide sufficiently-accurate and clinically-acceptable estimates of glucose levels for timely, proactive therapy and should be considered as the modeling engine for predictive monitoring of patients with type 1 diabetes mellitus. It also suggests that AR models can be made portable from individual to individual with minor performance penalties, while greatly reducing the burden associated with model tuning and data collection for model development.

Glucose control in pediatric intensive care unit patients using an insulin-glucose algorithm

BACKGROUND

Control of hyperglycemia in adult medical and surgical intensive care units (ICUs) has been shown to dramatically decrease morbidity and mortality. Algorithms to achieve glycemic control in the ICU setting are evolving. We have evaluated the use of a discrete proportional-integral-derivative (PID) algorithm to control hyperglycemia in pediatric ICU (PICU) patients both with and without diabetes.

METHODS

Six PICU patients [four with diabetic ketoacidosis (DKA) and two with glucocorticoid-induced hyperglycemia] with glucose values >150 mg/dL were enrolled. Their hyperglycemia was managed with a PID algorithm that provided recommendations for both changes in the intravenous insulin infusion rate and the time to obtain the next discrete glucose value. Glucose targets were adjusted based on clinical circumstances.

RESULTS

Patients (mean age 9.2 years; range 1.8-14 years) utilized the algorithm for a total of 454.4 h. Mean time to the initial glucose target was 8.7 h (range 1.3-15.1 h) in five patients. One subject with hyperosmolar DKA did not achieve target before discharge from the PICU, and another was at target when the algorithm was initiated. After the glucose target was achieved, the mean SD was 23.5 mg/dL, and glucose values were >40 mg/dL above target 13% of the time and <40 mg/dL below target 1% of the time. There were no glucose values <55 mg/dL.

CONCLUSION

The PID algorithm safely and effectively controlled hyperglycemia in a PICU, despite multiple changes in intravenous fluids, steroid doses (including high-dose pulses), and hemodialysis.

[Sub-cutaneous closed-loop: continuous glucose measurement and external insulin pump: hope in treatment of type 1 diabetes]

The development of an artificial pancreas allowing a continuous insulin infusion according to glucose measurement is closed to be an ideal device for type 1 diabetic patients and for the diabetologits. The development of miniaturized external pumps infusing short acting analogues with pharmacokinetic profile closer to physiology, and the availability of accurate continuous glucose sensors has led to the development of closed-loop subcutaneous system. The feasibility of this solution as been proven at a small scale but remains to be confirmed in a home setting. Intermediate solutions, such as semi-automatic systems might be immediately valuable.

GIM, simulation software of meal glucose-insulin model

BACKGROUND

A simulation model of the glucose-insulin system in normal life conditions can be very useful in diabetes research, e.g., testing insulin infusion algorithms and decision support systems and assessing glucose sensor performance and patient and student training. A new meal simulation model has been proposed that incorporates state-of-the-art quantitative knowledge on glucose metabolism and its control by insulin at both organ/tissue and whole-body levels. This article presents the interactive simulation software GIM (glucose insulin model), which implements this model.

METHODS

The model is implemented in MATLAB, version 7.0.1, and is designed with a windows interface that allows the user to easily simulate a 24-hour daily life of a normal, type 2, or type 1 diabetic subject. A Simulink version is also available. Three meals a day are considered. Both open- and closed-loop controls are available for simulating a type 1 diabetic subject.

RESULTS

Software options are described in detail. Case studies are presented to illustrate the potential of the software, e.g., compare a normal subject vs an insulin-resistant subject or open-loop vs closed-loop insulin infusion in type 1 diabetes treatment.

CONCLUSIONS

User-friendly software that implements a state-of-the-art physiological model of the glucose-insulin system during a meal has been presented. The GIM graphical interface makes its use extremely easy for investigators without specific expertise in modeling.

A "pancreatic tooth" design best accommodates the limitations of current artificial pancreas technology

Inadequately responsive glycaemic control is an important factor in the causation of diabetic end-organ damage. Artificial or hybrid bioartificial pancreases can provide responsive glycaemic control that can reduce the enormous personal suffering and socio-economic costs of diabetes. However, they share the shortcomings of limited operational life, due to depletion of stores or failure of component parts. A pancreatic tooth design provides accessibility for the purposes of replenishment or replacement. In addition, the mouth also provides a sheltered location, is more resistant to diabetic changes and less prone to thermoregulatory changes than subcutaneous tissues, and is adapted to cope with the occasional pathogen load. The device would consist of two parts: a permanent implant with an angiogenic capillary plexus that is the blood contacting interface and a crown containing the artificial or bio-artificial pancreatic systems: the accessibility of which confers ease of replenishment and replacement, among other advantages.

Recent progress in analytical instrumentation for glycemic control in diabetic and critically ill patients

Implementing strict glycemic control can reduce the risk of serious complications in both diabetic and critically ill patients. For this reason, many different analytical, mainly electrochemical and optical sensor approaches for glucose measurements have been developed. Self-monitoring of blood glucose (SMBG) has been recognised as being an indispensable tool for intensive diabetes therapy. Recent progress in analytical instrumentation, allowing submicroliter samples of blood, alternative site testing, reduced test time, autocalibration, and improved precision, is comprehensively described in this review. Continuous blood glucose monitoring techniques and insulin infusion strategies, developmental steps towards the realization of the dream of an artificial pancreas under closed loop control, are presented. Progress in glucose sensing and glycemic control for both patient groups is discussed by assessing recent published literature (up to 2006). The state-of-the-art and trends in analytical techniques (either episodic, intermittent or continuous, minimal-invasive, or noninvasive) detailed in this review will provide researchers, health professionals and the diabetic community with a comprehensive overview of the potential of next-generation instrumentation suited to either short- and long-term implantation or ex vivo measurement in combination with appropriate body interfaces such as microdialysis catheters.

The artificial pancreas: how sweet engineering will solve bitter problems

An artificial pancreas is a closed-loop system containing only synthetic materials which substitutes for an endocrine pancreas. No artificial pancreas system is currently approved; however, devices that could become components of such a system are now becoming commercially available. An artificial pancreas will consist of functionally integrated components that will continuously sense glucose levels, determine appropriate insulin dosages, and deliver the insulin. Any proposed closed loop system will be closely scrutinized for its safety, efficacy, and economic impact. Closed loop control utilizes models of glucose homeostasis which account for the influences of feeding, stress, insulin, exercise, and other factors on blood glucose levels. Models are necessary for understanding the relationship between blood glucose levels and insulin dosing; developing algorithms to control insulin dosing; and customizing each user's system based on individual responses to factors that influence glycemia. Components of an artificial pancreas are now being developed, including continuous glucose sensors; insulin pumps for parenteral delivery; and control software, all linked through wireless communication systems. Although a closed-loop system providing glucagon has not been reported in 40 years, the use of glucagon to prevent hypoglycemia is physiologically attractive and future devices might utilize this hormone. No demonstration of long-term closed loop control of glucose in a free-living human with diabetes has been reported to date, but many centers around the world are working on closed loop control systems. It is expected that many types of artificial pancreas systems will eventually be available, and they will greatly benefit patients with diabetes.

Feasibility of automating insulin delivery for the treatment of type 1 diabetes

An automated closed-loop insulin delivery system based on subcutaneous glucose sensing and subcutaneous insulin delivery was evaluated in 10 subjects with type 1 diabetes (2 men, 8 women, mean [+/-SD] age 43.4 +/- 11.4 years, duration of diabetes 18.2 +/- 13.5 years). Closed-loop control was assessed over approximately 30 h and compared with open-loop control assessed over 3 days. Closed-loop insulin delivery was calculated using a model of the beta-cell's multiphasic insulin response to glucose. Plasma glucose was 160 +/- 66 mg/dl at the start of closed loop and was thereafter reduced to 71 +/- 19 by 1:00 p.m. (preprandial lunch). Fasting glucose the subsequent morning on closed loop was not different from target (124 +/- 25 vs. 120 mg/dl, respectively; P > 0.05). Mean glucose levels were not different between the open and closed loop (133 +/- 63 vs. 133 +/- 52 mg/dl, respectively; P > 0.65). However, glucose was within the range 70-180 mg/dl 75% of the time under closed loop versus 63% for open loop. Incidence of biochemical hypoglycemia (blood glucose <60 mg/dl) was similar under the two treatments. There were no episodes of severe hypoglycemia. The data provide proof of concept that glycemic control can be achieved by a completely automated external closed-loop insulin delivery system.

Continuous glucose monitoring and external insulin pump: towards a subcutaneous closed loop

The development of an artificial pancreas for the treatment of type 1 diabetes is a highly desired endeavour for type 1 diabetic patients, physicians, scientists and engineers. The development of the first miniaturized external pumps in the 70s and the pharmacokinetic properties of short acting insulin analogues, closer to physiology, have raised hopes for the elaboration of such a device. Recent technological progress in the development of continuous glucose sensors, have improved the reliability and accuracy of these devices. This has led to the development of prototypes of closed-loop system based on the combination of a continuous monitor, a control algorithm, and an insulin pump. This review focuses on the SC-SC approach, employing subcutaneous glucose monitoring and subcutaneous insulin delivery. The feasibility of this solution has been proven at a small scale, but remains to be confirmed in the home setting. Intermediate solutions, such as semi-automatic systems, might be immediately valuable.

Application of Micro- and Nano-Electromechanical Devices to Drug Delivery

Micro- and nano-electromechanical systems (MEMS and NEMS)-based drug delivery devices have become commercially-feasible due to converging technologies and regulatory accommodation. The FDA Office of Combination Products coordinates review of innovative medical therapies that join elements from multiple established categories: drugs, devices, and biologics. Combination products constructed using MEMS or NEMS technology offer revolutionary opportunities to address unmet medical needs related to dosing. These products have the potential to completely control drug release, meeting requirements for on-demand pulsatile or adjustable continuous administration for extended periods. MEMS or NEMS technologies, materials science, data management, and biological science have all significantly developed in recent years, providing a multidisciplinary foundation for developing integrated therapeutic systems. If small-scale biosensor and drug reservoir units are combined and implanted, a wireless integrated system can regulate drug release, receive sensor feedback, and transmit updates. For example, an "artificial pancreas" implementation of an integrated therapeutic system would improve diabetes management. The tools of microfabrication technology, information science, and systems biology are being combined to design increasingly sophisticated drug delivery systems that promise to significantly improve medical care.