Inpatient trial of an artificial pancreas based on multiple model probabilistic predictive control with repeated large unannounced meals

Multivariable Automated Insulin Delivery System for Handling Planned and Spontaneous Physical Activities

BACKGROUND

Hybrid closed-loop control of glucose levels in people with type 1 diabetes mellitus (T1D) is limited by the requirements on users to manually announce physical activity (PA) and meals to the artificial pancreas system. Multivariable automated insulin delivery (mvAID) systems that can handle unannounced PAs and meals without any manual announcements by the user can improve glycemic control by modulating insulin dosing in response to the occurrence and intensity of spontaneous physical activities.

METHODS

An mvAID system is developed to supplement the glucose measurements with additional physiological signals from a wristband device, with the signals analyzed using artificial intelligence algorithms to automatically detect the occurrence of PA and estimate its intensity. This additional information gained from the physiological signals enables more proactive insulin dosing adjustments in response to both planned exercise and spontaneous unanticipated physical activities.

RESULTS

In silico studies of the mvAID illustrate the safety and efficacy of the system. The mvAID is translated to pilot clinical studies to assess its performance, and the clinical experiments demonstrate an increased time in range and reduced risk of hypoglycemia following unannounced PA and meals.

CONCLUSIONS

The mvAID systems can increase the safety and efficacy of insulin delivery in the presence of unannounced physical activities and meals, leading to improved lives and less burden on people with T1D.

Diabetes Technology Meeting 2021

Diabetes Technology Society hosted its annual Diabetes Technology Meeting on November 4 to November 6, 2021. This meeting brought together speakers to discuss various developments within the field of diabetes technology. Meeting topics included blood glucose monitoring, continuous glucose monitoring, novel sensors, direct-to-consumer telehealth, metrics for glycemia, software for diabetes, regulation of diabetes technology, diabetes data science, artificial pancreas, novel insulins, insulin delivery, skin trauma, metabesity, precision diabetes, diversity in diabetes technology, use of diabetes technology in pregnancy, and green diabetes. A live demonstration on a mobile app to monitor diabetic foot wounds was presented.

Closed-Loop Artificial Pancreas Therapy for Type 1 Diabetes

PURPOSE OF REVIEW

Closed-loop insulin pump systems (artificial pancreas) represent the cutting edge of insulin delivery technology. There are only a few systems currently approved for use in the USA: the MiniMed 670G/770G (which share an algorithm), t:slim X2 Control IQ, and the Omnipod 5. We review these systems and look into the future of the technology.

RECENT FINDINGS

All of the approved closed-loop insulin pump systems have demonstrated in multicenter prospective trials improvements in time in range, hemoglobin A1c, and time spent in hypoglycemia. The newer systems have also improved time spent in automation. Comparisons between the systems with regard to glycemic control are difficult to make due to differences in clinical trial design, but there are notable differences in the user experience between systems. The past few years have been a time of exponential development in the field of closed-loop insulin pump systems. However, more research is needed to provide full automation of these systems without any need for information from the user.

Fault Tolerant Strategies for Automated Insulin Delivery Considering the Human Component: Current and Future Perspectives

Automated Insulin Delivery (AID) are systems developed for daily use by people with type 1 diabetes (T1D). To ensure the safety of users, it is essential to consider how the human factor affects the performance and safety of these devices. While there are numerous publications on hardware-related failures of AID systems, there are few studies on the human component of the system. From a control point of view, people with T1D using AID systems are at the same time the plant to be controlled and the plant operator. Therefore, users may induce faults in the controller, sensors, actuators, and the plant itself. Strategies to cope with the human interaction in AID systems are needed for further development of the technology. In this paper, we present an analysis of potential faults introduced by AID users when the system is under normal operation. This is followed by a review of current fault tolerant control (FTC) approaches to identify missing areas of research. The paper concludes with a discussion on future directions for the new generation of FTC AID systems.

A Hybrid Automata Approach for Monitoring the Patient in the Loop in Artificial Pancreas Systems

The use of automated insulin delivery systems has become a reality for people with type 1 diabetes (T1D), with several hybrid systems already on the market. One of the particularities of this technology is that the patient is in the loop. People with T1D are the plant to control and also a plant operator, because they may have to provide information to the control loop. The most immediate information provided by patients that affects performance and safety are the announcement of meals and exercise. Therefore, to ensure safety and performance, the human factor impact needs to be addressed by designing fault monitoring strategies. In this paper, a monitoring system is developed to diagnose potential patient modes and faults. The monitoring system is based on the residual generation of a bank of observers. To that aim, a linear parameter varying (LPV) polytopic representation of the system is adopted and a bank of Kalman filters is designed using linear matrix inequalities (LMI). The system uncertainty is propagated using a zonotopic-set representation, which allows determining confidence bounds for each of the observer outputs and residuals. For the detection of modes, a hybrid automaton model is generated and diagnosis is performed by interpreting the events and transitions within the automaton. The developed system is tested in simulation, showing the potential benefits of using the proposed approach for artificial pancreas systems.

In-Silico Evaluation of Glucose Regulation Using Policy Gradient Reinforcement Learning for Patients with Type 1 Diabetes Mellitus

In this paper, we test and evaluate policy gradient reinforcement learning for automated blood glucose control in patients with Type 1 Diabetes Mellitus. Recent research has shown that reinforcement learning is a promising approach to accommodate the need for individualized blood glucose level control algorithms. The motivation for using policy gradient algorithms comes from the fact that adaptively administering insulin is an inherently continuous task. Policy gradient algorithms are known to be superior in continuous high-dimensional control tasks. Previously, most of the approaches for automated blood glucose control using reinforcement learning has used a finite set of actions. We use the Trust-Region Policy Optimization algorithm in this work. It represents the state of the art for deep policy gradient algorithms. The experiments are carried out in-silico using the Hovorka model, and stochastic behavior is modeled through simulated carbohydrate counting errors to illustrate the full potential of the framework. Furthermore, we use a model-free approach where no prior information about the patient is given to the algorithm. Our experiments show that the reinforcement learning agent is able to compete with and sometimes outperform state-of-the-art model predictive control in blood glucose regulation.

Safety and Efficacy of Initializing the Control-IQ Artificial Pancreas System Based on Total Daily Insulin in Adolescents with Type 1 Diabetes

Objective: To assess the safety and efficacy of a simplified initialization for the Tandem t:slim X2 Control-IQ hybrid closed-loop system, using parameters based on total daily insulin ("MyTDI") in adolescents with type 1 diabetes under usual activity and during periods of increased exercise. Research Design and Methods: Adolescents with type 1 diabetes 12-18 years of age used Control-IQ for 5 days at home using their usual parameters. Upon arrival at a 60-h ski camp, participants were randomized to either continue Control-IQ using their home settings or to reinitialize Control-IQ with MyTDI parameters. Control-IQ use continued for 5 days following camp. The effect of MyTDI on continuous glucose monitoring outcomes were analyzed using repeated measures analysis of variance (ANOVA): baseline, camp, and at home. Results: Twenty participants were enrolled and completed the study; two participants were excluded from the analysis due to absence from ski camp (1) and illness (1). Time in range was similar between both groups at home and camp. A tendency to higher time <70 mg/dL in the MyTDI group was present but only during camp (median 3.8% vs. 1.4%, P = 0.057). MyTDI users with bolus/TDI ratios >40% tended to show greater time in the euglycemic range improvements between baseline and home than users with ratios <40% (+16.3% vs. -9.0%, P = 0.012). All participants maintained an average of 95% time in closed loop (84.1%-100%). Conclusions: MyTDI is a safe, effective, and easy way to determine insulin parameters for use in the Control-IQ artificial pancreas. Future modifications to account for the influence of carbohydrate intake on MyTDI calculations might further improve time in range.

Glucose outcomes of a learning-type artificial pancreas with an unannounced meal in type 1 diabetes

BACKGROUND AND OBJECTIVES

Glycemic control with unannounced meals is the major challenge for artificial pancreas. In this study, we described the performance and safety of learning-type model predictive control (L-MPC) for artificial pancreas challenged by an unannounced meal in type 1 diabetes (T1D).

METHODS

This closed-loop (CL) system was tested in 29 T1D patients at one site in a 4 h inpatient open-label study. Participants used an L-MPC CL system for 6 days after 2-day system identification using open-loop (OL) insulin system. During the CL period, the L-MPC system was started from 8:00 am to noon each day. At 9:00 am, each participant consumed 50 g of carbohydrates with no prandial insulin bolus. At 9:30 am on CL-Day 4 or CL-Day 6, participants rode bicycles for 20 minutes or drank 50 ml of beer, in a random order.

RESULTS

As the primary outcome, TIR on CL-Day 3 was 65.2±23.3%, which was 9.8 points higher (95% CI 1.8 to 17.8; P = 0.019) than that on CL-Day 1. The time of glucose >10 mmol/L was decreased by 11.0% (95% CI -18.7 to 3.3; P = 0.007), and mean glucose level was decreased by 1.1 mmol/L (95% CI -1.1 to 0.5; P = 0.000). The total daily insulin dosage showed no significant difference (-0.1U, 95% CI -1.34 to 1.32; P = 0.982). Compared with OL-Day1 with a postprandial bolus, the TIR was increased by 13.7 points (95% CI 1.4 to 26.0; P = 0.030), the time of glucose >10 mmol/L and the mean glucose level were also decreased. Compared with the exercise day (CL-Day E, 62.0 ± 23.3%; P = 0.347) or alcohol day (CL-Day A, 64.0 ± 23.6%; P = 0.756), there was no statistically significant difference in terms of TIR, time of glucose >10 mmol/L and mean glucose level. No severe hypoglycemic events occurred and hypoglycemic episodes were not increased by using closed-loop insulin system.

CONCLUSION

The L-MPC CL insulin system achieved good glycemic control challenged by an unannounced meal.

Efficacy of the Diabeloop closed-loop system to improve glycaemic control in patients with type 1 diabetes exposed to gastronomic dinners or to sustained physical exercise

AIMS

To compare closed-loop (CL) and open-loop (OL) systems for glycaemic control in patients with type 1 diabetes (T1D) exposed to real-life challenging situations (gastronomic dinners or sustained physical exercise).

METHODS

Thirty-eight adult patients with T1D were included in a three-armed randomized pilot trial (Diabeloop WP6.2 trial) comparing glucose control using a CL system with use of an OL device during two crossover 72-hour periods in one of the three following situations: large (gastronomic) dinners; sustained and repeated bouts of physical exercise (with uncontrolled food intake); or control (rest conditions). Outcomes included time in spent in the glucose ranges of 4.4-7.8 mmol/L and 3.9-10.0 mmol/L, and time in hypo- and hyperglycaemia.

RESULTS

Time spent overnight in the tight range of 4.4 to 7.8 mmol/L was longer with CL (mean values: 63.2% vs 40.9% with OL; P ≤ .0001). Time spent during the day in the range of 3.9 to 10.0 mmol/L was also longer with CL (79.4% vs 64.1% with OL; P ≤ .0001). Participants using the CL system spent less time during the day with hyperglycaemic excursions (glucose >10.0 mmol/L) compared to those using an OL system (17.9% vs 31.9%; P ≤ .0001), and the proportions of time spent during the day with hyperglycaemic excursions of those using the CL system in the gastronomic dinner and physical exercise subgroups were of similar magnitude to those in the control subgroup (18.1 ± 6.3%, 17.2 ± 8.1% and 18.4 ± 12.5%, respectively). Finally, times spent in hypoglycaemia were short and not significantly different among the groups.

CONCLUSIONS

The Diabeloop CL system is superior to OL devices in reducing hyperglycaemic excursions in patients with T1D exposed to gastronomic dinners, or exposed to physical exercise followed by uncontrolled food and carbohydrate intake.

Chatteringfree hybrid adaptive neuro-fuzzy inference system-particle swarm optimisation data fusion-based BG-level control

In this study, a closed-loop control scheme is proposed for the glucose-insulin regulatory system in type-1 diabetic mellitus (T1DM) patients. Some innovative hybrid glucose-insulin regulators have combined artificial intelligence such as fuzzy logic and genetic algorithm with well known Palumbo model to regulate the blood glucose (BG) level in T1DM patients. However, most of these approaches have focused on the glucose reference tracking, and the qualitative of this tracking such as chattering reduction of insulin injection has not been well-studied. Higher-order sliding mode (HoSM) controllers have been employed to attenuate the effect of chattering. Owing to the delayed nature and non-linear property of glucose-insulin mechanism as well as various unmeasurable disturbances, even the HoSM methods are partly successful. In this study, data fusion of adaptive neuro-fuzzy inference systems optimised by particle swarm optimisation has been presented. The excellent performance of the proposed hybrid controller, i.e. desired BG-level tracking and chattering reduction in the presence of daily glucose-level disturbances is verified.

Realizing a Closed-Loop (Artificial Pancreas) System for the Treatment of Type 1 Diabetes

Recent, rapid changes in the treatment of type 1 diabetes have allowed for commercialization of an "artificial pancreas" that is better described as a closed-loop controller of insulin delivery. This review presents the current state of closed-loop control systems and expected future developments with a discussion of the human factor issues in allowing automation of glucose control. The goal of these systems is to minimize or prevent both short-term and long-term complications from diabetes and to decrease the daily burden of managing diabetes. The closed-loop systems are generally very effective and safe at night, have allowed for improved sleep, and have decreased the burden of diabetes management overnight. However, there are still significant barriers to achieving excellent daytime glucose control while simultaneously decreasing the burden of daytime diabetes management. These systems use a subcutaneous continuous glucose sensor, an algorithm that accounts for the current glucose and rate of change of the glucose, and the amount of insulin that has already been delivered to safely deliver insulin to control hyperglycemia, while minimizing the risk of hypoglycemia. The future challenge will be to allow for full closed-loop control with minimal burden on the patient during the day, alleviating meal announcements, carbohydrate counting, alerts, and maintenance. The human factors involved with interfacing with a closed-loop system and allowing the system to take control of diabetes management are significant. It is important to find a balance between enthusiasm and realistic expectations and experiences with the closed-loop system.

Importance of inpatient hypoglycaemia: impact, prediction and prevention

Hypoglycaemia is a key barrier to achieving euglycaemic control in people who are hospitalized. Inpatient hypoglycaemia has been linked to adverse clinical outcomes, including mortality and longer stay in hospital. A number of studies have applied mathematical tools and statistical models to predict inpatient hypoglycaemia and identify factors that may result in hypoglycaemic events. Several different approaches have been tested to prevent inpatient hypoglycaemia. These can be categorized as human intervention, computerized methods or application of medical devices. In this review we provide an overview of the epidemiology of inpatient hypoglycaemia and its impact on patients and hospitals. We also discuss the existing methodology used to predict inpatient hypoglycaemia and the limited number of trials performed to prevent inpatient hypoglycaemia. The review highlights the urgent need for evidence-based methods to reduce inpatient hypoglycaemia.

Measurement Performance of Two Continuous Tissue Glucose Monitoring Systems Intended for Replacement of Blood Glucose Monitoring

BACKGROUND

Currently, two systems for continuous tissue glucose monitoring (CGM) (Dexcom^® G5 [DG5] and FreeStyle Libre [FL]) are intended to replace blood glucose monitoring (BGM) and, according to manufacturer labeling, are distributed as such in some jurisdictions, including the United States and the European Union.

METHODS

The measurement performance of these two systems in comparison with a BGM system was analyzed in a 14-day study with 20 participants comprising study site visits, which included phases of induced rapid glucose changes, and home use phases. Performance analysis was mainly based on deviations between CGM readings and BGM results. Sensor-to-sensor precision was also analyzed.

RESULTS

Approximately 25% of DG5 and FL results showed differences from BGM results exceeding 15 mg/dL or 15% (at glucose concentration below or above 100 mg/dL, respectively) at times of therapeutic decisions, and ∼5% of differences exceeded 30 mg/dL or 30%. Performance was different depending on the setting (study site visits, home use phases, and phases of induced rapid glucose changes). In consensus error grid (CEG) analysis, both systems showed >99.5% of results within the clinically acceptable zones A and B.

CONCLUSIONS

In this study, both systems showed deviations from blood glucose (BG) measurements, the current standard approach in diabetes therapy. Although a large percentage of results was found in CEG zones A and B, for approximately one in four therapeutic decisions, CGM and BG readings differed by at least 15 mg/dL or 15%. Such deviations should be taken into account when using CGM systems.

Fully Closed-Loop Multiple Model Probabilistic Predictive Controller Artificial Pancreas Performance in Adolescents and Adults in a Supervised Hotel Setting

BACKGROUND

Initial Food and Drug Administration-approved artificial pancreas (AP) systems will be hybrid closed-loop systems that require prandial meal announcements and will not eliminate the burden of premeal insulin dosing. Multiple model probabilistic predictive control (MMPPC) is a fully closed-loop system that uses probabilistic estimation of meals to allow for automated meal detection. In this study, we describe the safety and performance of the MMPPC system with announced and unannounced meals in a supervised hotel setting.

RESEARCH DESIGN AND METHODS

The Android phone-based AP system with remote monitoring was tested for 72 h in six adults and four adolescents across three clinical sites with daily exercise and meal challenges involving both three announced (manual bolus by patient) and six unannounced (no bolus by patient) meals. Safety criteria were predefined. Controller aggressiveness was adapted daily based on prior hypoglycemic events.

RESULTS

Mean 24-h continuous glucose monitor (CGM) was 157.4 ± 14.4 mg/dL, with 63.6 ± 9.2% of readings between 70 and 180 mg/dL, 2.9 ± 2.3% of readings <70 mg/dL, and 9.0 ± 3.9% of readings >250 mg/dL. Moderate hyperglycemia was relatively common with 24.6 ± 6.2% of readings between 180 and 250 mg/dL, primarily within 3 h after a meal. Overnight mean CGM was 139.6 ± 27.6 mg/dL, with 77.9 ± 16.4% between 70 and 180 mg/dL, 3.0 ± 4.5% <70 mg/dL, 17.1 ± 14.9% between 180 and 250 mg/dL, and 2.0 ± 4.5%> 250 mg/dL. Postprandial hyperglycemia was more common for unannounced meals compared with announced meals (4-h postmeal CGM 197.8 ± 44.1 vs. 140.6 ± 35.0 mg/dL; P < 0.001). No participants met safety stopping criteria.

CONCLUSIONS

MMPPC was safe in a supervised setting despite meal and exercise challenges. Further studies are needed in a less supervised environment.

Real-Time Detection of Infusion Site Failures in a Closed-Loop Artificial Pancreas

BACKGROUND

As evidence emerges that artificial pancreas systems improve clinical outcomes for patients with type 1 diabetes, the burden of this disease will hopefully begin to be alleviated for many patients and caregivers. However, reliance on automated insulin delivery potentially means patients will be slower to act when devices stop functioning appropriately. One such scenario involves an insulin infusion site failure, where the insulin that is recorded as delivered fails to affect the patient's glucose as expected. Alerting patients to these events in real time would potentially reduce hyperglycemia and ketosis associated with infusion site failures.

METHODS

An infusion site failure detection algorithm was deployed in a randomized crossover study with artificial pancreas and sensor-augmented pump arms in an outpatient setting. Each arm lasted two weeks. Nineteen participants wore infusion sets for up to 7 days. Clinicians contacted patients to confirm infusion site failures detected by the algorithm and instructed on set replacement if failure was confirmed.

RESULTS

In real time and under zone model predictive control, the infusion site failure detection algorithm achieved a sensitivity of 88.0% (n = 25) while issuing only 0.22 false positives per day, compared with a sensitivity of 73.3% (n = 15) and 0.27 false positives per day in the SAP arm (as indicated by retrospective analysis). No association between intervention strategy and duration of infusion sets was observed ( P = .58).

CONCLUSIONS

As patient burden is reduced by each generation of advanced diabetes technology, fault detection algorithms will help ensure that patients are alerted when they need to manually intervene. Clinical Trial Identifier: www.clinicaltrials.gov,NCT02773875.

Automatic Detection and Estimation of Unannounced Meals for Multivariable Artificial Pancreas System

BACKGROUND

Automatically attenuating the postprandial rise in the blood glucose concentration without manual meal announcement is a significant challenge for artificial pancreas (AP) systems. In this study, a meal module is proposed to detect the consumption of a meal and to estimate the amount of carbohydrate (CHO) intake.

METHODS

The meals are detected based on qualitative variables describing variation of continuous glucose monitoring (CGM) readings. The CHO content of the meals/snacks is estimated by a fuzzy system using CGM and subcutaneous insulin delivery data. The meal bolus amount is computed according to the patient's insulin to CHO ratio. Integration of the meal module into a multivariable AP system allows revision of estimated CHO based on knowledge about physical activity, sleep, and the risk of hypoglycemia before the final decision for a meal bolus is made.

RESULTS

The algorithm is evaluated by using 117 meals/snacks in retrospective data from 11 subjects with type 1 diabetes. Sensitivity, defined as the percentage of correctly detected meals and snacks, is 93.5% for meals and 68.0% for snacks. The percentage of false positives, defined as the proportion of false detections relative to the total number of detected meals and snacks, is 20.8%.

CONCLUSIONS

Integration of a meal detection module in an AP system is a further step toward an automated AP without manual entries. Detection of a consumed meal/snack and infusion of insulin boluses using an estimate of CHO enables the AP system to automatically prevent postprandial hyperglycemia.

The challenges of achieving postprandial glucose control using closed-loop systems in patients with type 1 diabetes

For patients with type 1 diabetes, closed-loop delivery systems (CLS) combining an insulin pump, a glucose sensor and a dosing algorithm allowing a dynamic hormonal infusion have been shown to improve glucose control when compared with conventional therapy. Yet, reducing glucose excursion and simplification of prandial insulin doses remain a challenge. The objective of this literature review is to examine current meal-time strategies in the context of automated delivery systems in adults and children with type 1 diabetes. Current challenges and considerations for post-meal glucose control will also be discussed. Despite promising results with meal detection, the fully automated CLS has yet failed to provide comparable glucose control to CLS with carbohydrate-matched bolus in the post-meal period. The latter strategy has been efficient in controlling post-meal glucose using different algorithms and in various settings, but at the cost of a meal carbohydrate counting burden for patients. Further improvements in meal detection algorithms or simplified meal-priming boluses may represent interesting avenues. The greatest challenges remain in regards to the pharmacokinetic and dynamic profiles of available rapid insulins as well as sensor accuracy and lag-time. New and upcoming faster acting insulins could provide important benefits. Multi-hormone CLS (eg, dual-hormone combining insulin with glucagon or pramlintide) and adjunctive therapy (eg, GLP-1 and SGLT2 inhibitors) also represent promising options. Meal glucose control with the artificial pancreas remains an important challenge for which the optimal strategy is still to be determined.

Twelve-Week 24/7 Ambulatory Artificial Pancreas With Weekly Adaptation of Insulin Delivery Settings: Effect on Hemoglobin A1c and Hypoglycemia

OBJECTIVE

Artificial pancreas (AP) systems are best positioned for optimal treatment of type 1 diabetes (T1D) and are currently being tested in outpatient clinical trials. Our consortium developed and tested a novel adaptive AP in an outpatient, single-arm, uncontrolled multicenter clinical trial lasting 12 weeks.

RESEARCH DESIGN AND METHODS

Thirty adults with T1D completed a continuous glucose monitor (CGM)-augmented 1-week sensor-augmented pump (SAP) period. After the AP was started, basal insulin delivery settings used by the AP for initialization were adapted weekly, and carbohydrate ratios were adapted every 4 weeks by an algorithm running on a cloud-based server, with automatic data upload from devices. Adaptations were reviewed by expert study clinicians and patients. The primary end point was change in hemoglobin A_1c (HbA_1c). Outcomes are reported adhering to consensus recommendations on reporting of AP trials.

RESULTS

Twenty-nine patients completed the trial. HbA_1c, 7.0 ± 0.8% at the start of AP use, improved to 6.7 ± 0.6% after 12 weeks (-0.3, 95% CI -0.5 to -0.2, P < 0.001). Compared with the SAP run-in, CGM time spent in the hypoglycemic range improved during the day from 5.0 to 1.9% (-3.1, 95% CI -4.1 to -2.1, P < 0.001) and overnight from 4.1 to 1.1% (-3.1, 95% CI -4.2 to -1.9, P < 0.001). Whereas carbohydrate ratios were adapted to a larger extent initially with minimal changes thereafter, basal insulin was adapted throughout. Approximately 10% of adaptation recommendations were manually overridden. There were no protocol-related serious adverse events.

CONCLUSIONS

Use of our novel adaptive AP yielded significant reductions in HbA_1c and hypoglycemia.

Closed-Loop Control Without Meal Announcement in Type 1 Diabetes

OBJECTIVE

A fully closed-loop insulin-only system was developed to provide glucose control in patients with type 1 diabetes without requiring announcement of meals or activity. Our goal was to assess initial safety and efficacy of this system.

RESEARCH DESIGN AND METHODS

The multiple model probabilistic controller (MMPPC) anticipates meals when the patient is awake. The controller used the subject's basal rates and total daily insulin dose for initialization. The system was tested at two sites on 10 patients in a 30-h inpatient study, followed by 15 subjects at three sites in a 54-h supervised hotel study, where the controller was challenged by exercise and unannounced meals. The system was implemented on the UVA DiAs system using a Roche Spirit Combo Insulin Pump and a Dexcom G4 Continuous Glucose Monitor.

RESULTS

The mean overall (24-h basis) and nighttime (11 PM-7 AM) continuous glucose monitoring (CGM) values were 142 and 125 mg/dL during the inpatient study. The hotel study used a different daytime tuning and manual announcement, instead of automatic detection, of sleep and wake periods. This resulted in mean overall (24-h basis) and nighttime CGM values of 152 and 139 mg/dL for the hotel study and there was also a reduction in hypoglycemia events from 1.6 to 0.91 events/patient/day.

CONCLUSIONS

The MMPPC system achieved a mean glucose that would be particularly helpful for people with an elevated A1c as a result of frequent missed meal boluses. Current full closed loop has a higher risk for hypoglycemia when compared with algorithms using meal announcement.

Multivariable Adaptive Artificial Pancreas System in Type 1 Diabetes

PURPOSE OF REVIEW

The review summarizes the current state of the artificial pancreas (AP) systems and introduces various new modules that should be included in future AP systems.

RECENT FINDINGS

A fully automated AP must be able to detect and mitigate the effects of meals, exercise, stress and sleep on blood glucose concentrations. This can only be achieved by using a multivariable approach that leverages information from wearable devices that provide real-time streaming data about various physiological variables that indicate imminent changes in blood glucose concentrations caused by meals, exercise, stress and sleep. The development of a fully automated AP will necessitate the design of multivariable and adaptive systems that use information from wearable devices in addition to glucose sensors and modify the models used in their model-predictive alarm and control systems to adapt to the changes in the metabolic state of the user. These AP systems will also integrate modules for controller performance assessment, fault detection and diagnosis, machine learning and classification to interpret various signals and achieve fault-tolerant control. Advances in wearable devices, computational power, and safe and secure communications are enabling the development of fully automated multivariable AP systems.

Meal Detection and Carbohydrate Estimation Using Continuous Glucose Sensor Data

A meal detection and meal-size estimation algorithm is developed for use in artificial pancreas (AP) control systems for people with type 1 diabetes. The algorithm detects the consumption of a meal and estimates its carbohydrate (CHO) amount to determine the appropriate dose of insulin bolus for a meal. It can be used in AP systems without manual meal announcements, or as a safety feature for people who may forget entering meal information manually. Using qualitative representation of the filtered continuous glucose monitor signal, a time period labeled as meal flag is identified. At every sampling time during this time period, a fuzzy system estimates the amount of CHO. Meal size estimator uses both glucose sensor and insulin data. Meal insulin bolus is based on estimated CHO. The algorithm does not change the basal insulin rate. Thirty in silico subjects of the UVa/Padova simulator are used to illustrate the performance of the algorithm. For the evaluation dataset, the sensitivity and false positives detection rates are 91.3% and 9.3%, respectively, the absolute error in CHO estimation is 23.1%, the mean blood glucose level is 142 mg/dl, and glucose concentration stays in target range (70-180 mg/dl) for 76.8% of simulation duration on average.

Continuous Glucose Monitoring Enables the Detection of Losses in Infusion Set Actuation (LISAs)

Reliable continuous glucose monitoring (CGM) enables a variety of advanced technology for the treatment of type 1 diabetes. In addition to artificial pancreas algorithms that use CGM to automate continuous subcutaneous insulin infusion (CSII), CGM can also inform fault detection algorithms that alert patients to problems in CGM or CSII. Losses in infusion set actuation (LISAs) can adversely affect clinical outcomes, resulting in hyperglycemia due to impaired insulin delivery. Prolonged hyperglycemia may lead to diabetic ketoacidosis-a serious metabolic complication in type 1 diabetes. Therefore, an algorithm for the detection of LISAs based on CGM and CSII signals was developed to improve patient safety. The LISA detection algorithm is trained retrospectively on data from 62 infusion set insertions from 20 patients. The algorithm collects glucose and insulin data, and computes relevant fault metrics over two different sliding windows; an alarm sounds when these fault metrics are exceeded. With the chosen algorithm parameters, the LISA detection strategy achieved a sensitivity of 71.8% and issued 0.28 false positives per day on the training data. Validation on two independent data sets confirmed that similar performance is seen on data that was not used for training. The developed algorithm is able to effectively alert patients to possible infusion set failures in open-loop scenarios, with limited evidence of its extension to closed-loop scenarios.

Algorithms for a Single Hormone Closed-Loop Artificial Pancreas: Challenges Pertinent to Chemical Process Operations and Control

The development of a closed-loop artificial pancreas to regulate the blood glucose concentration of individuals with type 1 diabetes has been a focused area of research for over 50 years, with rapid progress during the past decade. The daily control challenges faced by someone with type 1 diabetes include asymmetric objectives and risks, and one-sided manipulated input action with frequent relatively fast disturbances. The major automation steps toward a closed-loop artificial pancreas include (i) monitoring and overnight alarms for hypoglycemia (low blood glucose); (ii) overnight low glucose suspend (LGS) systems to prevent hypoglycemia; and (iii) fully closed-loop systems that adjust insulin (and perhaps glucagon) to maintain desired blood glucose levels day and night. We focus on the steps that we used to develop and test a probabilistic, risk-based, model predictive control strategy for a fully closed-loop artificial pancreas. We complete the paper by discussing ramifications of lessons learned for chemical process systems applications.

Switched LPV Glucose Control in Type 1 Diabetes

OBJECTIVE

The purpose of this paper is to regulate the blood glucose level in Type 1 Diabetes Mellitus patients with a practical and flexible procedure that can switch among a finite number of distinct controllers, depending on the user's choice.

METHODS

A switched linear parameter-varying controller with multiple switching regions, related to hypo-, hyper-, and euglycemia situations, is designed. The key feature is to arrange the controller into a framework that provides stability and performance guaranty.

RESULTS

The closed-loop performance is tested on the complete in silico adult cohort of the UVA/Padova metabolic simulator, which has been accepted by the U.S. Food and Drug Administration in lieu of animal trials. The outcome produces comparable or improved results with respect to previous works.

CONCLUSION

The strategy is practical because it is based on a model tuned only with a priori patient information in order to cover the interpatient uncertainty. Results confirm that this control structure yields tangible improvements in minimizing risks of hyper- and hypoglycemia in scenarios with unannounced meals.

SIGNIFICANCE

This flexible procedure opens the possibility of taking into account, at the design stage, unannounced meals and/or patients' physical exercise.

Adjustment of Open-Loop Settings to Improve Closed-Loop Results in Type 1 Diabetes: A Multicenter Randomized Trial

CONTEXT

Closed-loop control (CLC) relies on an individual's open-loop insulin pump settings to initialize the system. Optimizing open-loop settings before using CLC usually requires significant time and effort.

OBJECTIVE

The objective was to investigate the effects of a one-time algorithmic adjustment of basal rate and insulin to carbohydrate ratio open-loop settings on the performance of CLC.

DESIGN

This study reports a multicenter, outpatient, randomized, crossover clinical trial.

PATIENTS

Thirty-seven adults with type 1 diabetes were enrolled at three clinical sites.

INTERVENTIONS

Each subject's insulin pump settings were subject to a one-time algorithmic adjustment based on 1 week of open-loop (i.e., home care) data collection. Subjects then underwent two 27-hour periods of CLC in random order with either unchanged (control) or algorithmic adjusted basal rate and carbohydrate ratio settings (adjusted) used to initialize the zone-model predictive control artificial pancreas controller. Subject's followed their usual meal-plan and had an unannounced exercise session.

MAIN OUTCOMES AND MEASURES

Time in the glucose range was 80-140 mg/dL, compared between both arms.

RESULTS

Thirty-two subjects completed the protocol. Median time in CLC was 25.3 hours. The median time in the 80-140 mg/dl range was similar in both groups (39.7% control, 44.2% adjusted). Subjects in both arms of CLC showed minimal time spent less than 70 mg/dl (median 1.34% and 1.37%, respectively). There were no significant differences more than 140 mg/dL.

CONCLUSIONS

A one-time algorithmic adjustment of open-loop settings did not alter glucose control in a relatively short duration outpatient closed-loop study. The CLC system proved very robust and adaptable, with minimal (<2%) time spent in the hypoglycemic range in either arm.

Efficacy of dual-hormone artificial pancreas to alleviate the carbohydrate-counting burden of type 1 diabetes: A randomized crossover trial

AIM

Carbohydrate-counting is a complex task for many patients with type 1 diabetes. This study examined whether an artificial pancreas, delivering insulin and glucagon based on glucose sensor readings, could alleviate the burden of carbohydrate-counting without degrading glucose control.

METHODS

Twelve adults were recruited into a randomized, three-way, crossover trial (ClinicalTrials.gov identifier No. NCT01930097). Participants were admitted on three occasions from 7AM to 9PM and consumed a low-carbohydrate breakfast (women: 30g; men: 50g), a medium-carbohydrate dinner (women: 50g; men: 70g) and a high-carbohydrate lunch (women: 90g; men: 120g). At each visit, glucose levels were randomly regulated by: (1) conventional pump therapy; (2) an artificial pancreas (AP) accompanied by prandial boluses, matching the meal's carbohydrate content based on insulin-to-carbohydrate ratios (AP with carbohydrate-counting); or (3) an AP accompanied by prandial boluses based on qualitative categorization (regular or large) of meal size (AP without carbohydrate-counting).

RESULTS

The AP without carbohydrate-counting achieved similar incremental AUC values compared with carbohydrate-counting after the low- (P=0.54) and medium- (P=0.38) carbohydrate meals, but yielded higher post-meal excursions after the high-carbohydrate meal (P=0.004). The AP with and without carbohydrate-counting yielded similar mean glucose levels (8.2±2.1mmol/L vs. 8.4±1.7mmol/L; P=0.52), and both strategies resulted in lower mean glucose compared with conventional pump therapy (9.6±2.0mmol/L; P=0.02 and P=0.03, respectively).

CONCLUSION

The AP with qualitative categorization of meal size could alleviate the burden of carbohydrate-counting without compromising glucose control, although more categories of meal sizes are probably needed to effectively control higher-carbohydrate meals.