Incorporating Unannounced Meals and Exercise in Adaptive Learning of Personalized Models for Multivariable Artificial Pancreas Systems

Recent advances in the precision control strategy of artificial pancreas

The scientific diagnosis and treatment of patients with diabetes require frequent blood glucose testing and insulin delivery to normoglycemia. Therefore, an artificial pancreas with a continuous blood glucose (BG) monitoring function is an urgent research target in the medical industry. The problem of closed-loop algorithmic control of the BG with a time delay is a key and difficult issue that needs to be overcome in the development of an artificial pancreas. Firstly, the composition, structure, and control characteristics of the artificial pancreas are introduced. Subsequently, the research progress of artificial pancreas control algorithms is reviewed, and the characteristics, advantages, and disadvantages of proportional-integral-differential control, model predictive control, and artificial intelligence control are compared and analyzed to determine whether they are suitable for the practical application of the artificial pancreas. Additionally, key advancements in areas such as blood glucose data monitoring, adaptive models, wearable devices, and fully automated artificial pancreas systems are also reviewed. Finally, this review highlights that meal prediction, control safety, integration, streamlining the optimization of control algorithms, constant temperature preservation of insulin, and dual-hormone artificial pancreas are issues that require further attention in the future.

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.

A practical approach based on learning-based model predictive control with minimal prior knowledge of patients for artificial pancreas

BACKGROUND AND OBJECTIVES

Complete identification of the glucose dynamics for a patient generally requires prior clinical procedures and several measurements for the patient. However, these steps may not be always feasible. To address this limitation, we propose a practical approach integrating learning-based model predictive control (MPC), adaptive basal and bolus injections, and suspension with minimal requirements of prior knowledge of the patient.

METHODS

The glucose dynamic system matrices were periodically updated using only input values, without any pretrained models. The optimal insulin dose was calculated based on a learning-based MPC algorithm. Meal detection and estimation modules were also introduced. The basal and bolus insulin injections were fine-tuned using the performance of glucose control from the previous day. To validate the proposed method, evaluations with 20 virtual patients from a type 1 diabetes metabolic simulator were employed.

RESULTS

Time-in-range (TIR) and time-below-range (TBR) were 90.8% (84.1% - 95.6%) and 0.3% (0% - 0.8%), as represented by the median, first (Q1), and third quartiles (Q3), respectively, when meal intakes were fully announced. When one out of three meal intake announcements was missing, TIR and TBR were 85.2% (75.0% - 88.9%) and 0.9% (0.4% - 1.1%), respectively.

CONCLUSIONS

The proposed approach obviates the need for prior tests from patients and shows effective regulation of blood glucose levels. From the perspective of practical implementation in clinical environments, to deal with minimal prior information of the patient, our study demonstrates how essential clinical knowledge and learning-based modules can be integrated into a control framework for an artificial pancreas.

Multi-Task Classification of Physical Activity and Acute Psychological Stress for Advanced Diabetes Treatment

Wearable sensor data can be integrated and interpreted to improve the treatment of chronic conditions, such as diabetes, by enabling adjustments in treatment decisions based on physical activity and psychological stress assessments. The challenges in using biological analytes to frequently detect physical activity (PA) and acute psychological stress (APS) in daily life necessitate the use of data from noninvasive sensors in wearable devices, such as wristbands. We developed a recurrent multi-task deep neural network (NN) with long-short-term-memory architecture to integrate data from multiple sensors (blood volume pulse, skin temperature, galvanic skin response, three-axis accelerometers) and simultaneously detect and classify the type of PA, namely, sedentary state, treadmill run, stationary bike, and APS, such as non-stress, emotional anxiety stress, mental stress, and estimate the energy expenditure (EE). The objective was to assess the feasibility of using the multi-task recurrent NN (RNN) rather than independent RNNs for detection and classification of AP and APS. The multi-task RNN achieves comparable performance to independent RNNs, with the multi-task RNN having F1 scores of 98.00% for PA and 98.97% for APS, and a root mean square error (RMSE) of 0.728 calhr.kg for EE estimation for testing data. The independent RNNs have F1 scores of 99.64% for PA and 98.83% for APS, and an RMSE of 0.666 calhr.kg for EE estimation. The results indicate that a multi-task RNN can effectively interpret the signals from wearable sensors. Additionally, we developed individual and multi-task extreme gradient boosting (XGBoost) for separate and simultaneous classification of PA types and APS types. Multi-task XGBoost achieved F1 scores of 99.89% and 98.31% for the classification of PA types and APS types, respectively, while the independent XGBoost achieved F1 scores of 99.68% and 96.77%, respectively. The results indicate that both multi-task RNN and XGBoost can be used for the detection and classification of PA and APS without loss of performance with respect to individual separate classification systems. People with diabetes can achieve better outcomes and quality of life by including physical activity and psychological stress assessments in treatment decision-making.

Near-optimal feedback control for postprandial glucose regulation in type 1 diabetes

This paper is focused on feedback control of postprandial glucose levels for patients with type 1 Diabetes Mellitus. There are two important limitations that make this a challenging problem. First, the slow subcutaneous insulin pharmacokinetics that introduces a significant lag into the control loop. Second, the positivity constraint on the control action, meaning that it is not possible to remove insulin from the body. In this paper, both issues are explicitly considered in the design process using the internal model control framework, to derive a near-optimal feedback controller. Optimality is understood here as minimizing the blood glucose peak after a meal intake and, at the same time, preventing glucose values below a prescribed threshold. It is shown how the proposed controller approaches the optimal closed-loop performance as a limit case. The theoretical results are supported by a numerical example and the feasibility of the overall strategy under uncertainties is illustrated using an extended version UVa/Padova metabolic simulator.

Strengths and Challenges of Closed-Loop Insulin Delivery During Exercise in People With Type 1 Diabetes: Potential Future Directions

Exercise has many physical and psychological benefits and is recommended for people with type 1 diabetes; however, there are many barriers to exercise, including glycemic instability and fear of hypoglycemia. Closed-loop (CL) systems have shown benefit in the overall glycemic management of type 1 diabetes, including improving HbA1c levels and reducing the incidence of nocturnal hypoglycemia; however, these systems are challenged by the rapidly changing insulin needs with exercise. This commentary focuses on the principles, strengths, and challenges of CL in the management of exercise, and discusses potential approaches, including the use of additional physiological signals, to address their shortcomings in the pursuit of fully automated CL systems.

Quantifying the impact of physical activity on future glucose trends using machine learning

Prevention of hypoglycemia (glucose <70 mg/dL) during aerobic exercise is a major challenge in type 1 diabetes. Providing predictions of glycemic changes during and following exercise can help people with type 1 diabetes avoid hypoglycemia. A unique dataset representing 320 days and 50,000 + time points of glycemic measurements was collected in adults with type 1 diabetes who participated in a 4-arm crossover study evaluating insulin-pump therapies, whereby each participant performed eight identically designed in-clinic exercise studies. We demonstrate that even under highly controlled conditions, there is considerable intra-participant and inter-participant variability in glucose outcomes during and following exercise. Participants with higher aerobic fitness exhibited significantly lower minimum glucose and steeper glucose declines during exercise. Adaptive, personalized machine learning (ML) algorithms were designed to predict exercise-related glucose changes. These algorithms achieved high accuracy in predicting the minimum glucose and hypoglycemia during and following exercise sessions, for all fitness levels.

Prior Informed Regularization of Recursively Updated Latent-Variables-Based Models with Missing Observations

Many data-driven modeling techniques identify locally valid, linear representations of time-varying or nonlinear systems, and thus the model parameters must be adaptively updated as the operating conditions of the system vary, though the model identification typically does not consider prior knowledge. In this work, we propose a new regularized partial least squares (rPLS) algorithm that incorporates prior knowledge in the model identification and can handle missing data in the independent covariates. This latent variable (LV) based modeling technique consists of three steps. First, a LV-based model is developed on the historical time series data. In the second step, the missing observations in the new incomplete data sample are estimated. Finally, the future values of the outputs are predicted as a linear combination of estimated scores and loadings. The model is recursively updated as new data are obtained from the system. The performance of the proposed rPLS and rPLS with exogenous inputs (rPLSX) algorithms are evaluated by modeling variations in glucose concentration (GC) of people with Type 1 diabetes (T1D) in response to meals and physical activities for prediction windows up to one hour, or 12 sampling instances, into the future. The proposed rPLS family of GC prediction models are evaluated with both in-silico and clinical experiment data and compared with the performance of recursive time series and kernel-based models. The root mean squared error (RMSE) with simulated subjects in the multivariable T1D simulator where physical activity effects are incorporated in GC variations are 2.52 and 5.81 mg/dL for 30 and 60 mins ahead predictions (respectively) when information for all meals and physical activities are used, increasing to 2.70 and 6.54 mg/dL (respectively) when meals and activities occurred, but the information is with-held from the modeling algorithms. The RMSE is 10.45 and 14.48 mg/dL for clinical study with prediction horizons of 30 and 60 mins, respectively. The low RMSE values demonstrate the effectiveness of the proposed rPLS approach compared to the conventional recursive modeling algorithms.

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 Multitask Learning Approach to Personalised Blood Glucose Prediction

Blood glucose prediction algorithms are key tools in the development of decision support systems and closed-loop insulin delivery systems for blood glucose control in diabetes. Deep learning models have provided leading results among machine learning algorithms to date in glucose prediction. However these models typically require large amounts of data to obtain best personalised glucose prediction results. Multitask learning facilitates an approach for leveraging data from multiple subjects while still learning accurate personalised models. In this work we present results comparing the effectiveness of multitask learning over sequential transfer learning, and learning only on subject-specific data with neural networks and support vector regression. The multitask learning approach shows consistent leading performance in predictive metrics at both short-term and long-term prediction horizons. We obtain a predictive accuracy (RMSE) of 18.8 2.3, 25.3 2.9, 31.8 3.9, 41.2 4.5, 47.2 4.6 mg/dL at 30, 45, 60, 90, and 120 min prediction horizons respectively, with at least 93\% clinically acceptable predictions using the Clarke Error Grid (EGA) at each prediction horizon. We also identify relevant prior information such as glycaemic variability that can be incorporated to improve predictive performance at long-term prediction horizons. Furthermore, we demonstrate consistent performance - 5% change in both RMSE and EGA (Zone A) - in rare cases of adverse glycaemic events with 1-6 weeks of training data. In conclusion, a multitask approach can allow for deploying personalised models even with significantly less subject-specific data without compromising performance.

Physical Activity and Psychological Stress Detection and Assessment of Their Effects on Glucose Concentration Predictions in Diabetes Management

OBJECTIVE

Continuous glucose monitoring (CGM) enables prediction of the future glucose concentration (GC) trajectory for making informed diabetes management decisions. The glucose concentration values are affected by various physiological and metabolic variations, such as physical activity (PA) and acute psychological stress (APS), in addition to meals and insulin. In this work, we extend our adaptive glucose modeling framework to incorporate the effects of PA and APS on the GC predictions.

METHODS

A wristband conducive of use by free-living ambulatory people is used. The measured physiological variables are analyzed to generate new quantifiable input features for PA and APS. Machine learning techniques estimate the type and intensity of the PA and APS when they occur individually and concurrently. Variables quantifying the characteristics of both PA and APS are integrated as exogenous inputs in an adaptive system identification technique for enhancing the accuracy of GC predictions. Data from clinical experiments illustrate the improvement in GC prediction accuracy.

RESULTS

The average mean absolute error (MAE) of one-hour-ahead GC predictions with testing data decreases from 35.1 to 31.9 mg/dL (p-value = 0.01) with the inclusion of PA information, and it decreases from 16.9 to 14.2 mg/dL (p-value = 0.006) with the inclusion of PA and APS information.

CONCLUSION

The first-ever glucose prediction model is developed that incorporates measures of physical activity and acute psychological stress to improve GC prediction accuracy.

SIGNIFICANCE

Modeling the effects of physical activity and acute psychological stress on glucose concentration values will improve diabetes management and enable informed meal, activity and insulin dosing decisions.

A New Stochastic Approach for Modeling Glycemic Disturbances in Type 2 Diabetes

OBJECTIVE

To improve insulin treatment in type 2 diabetes (T2D) using model-based control techniques, the underlying model needs to be individualized to each patient. Due to the impact of unknown meals, exercise and other factors on the blood glucose, it is difficult to utilize available data from continuous glucose monitors (CGMs) for model fitting and parameter estimation purposes.

METHODS

To overcome this problem, we propose a novel method for modeling the glycemic disturbances as a stochastic process. To differentiate meals from other glycemic disturbances, we model the meal intake as a separate stochastic process while encompassing all other disturbances in another stochastic process. Using particle filtering, we validate the model on simulations as well as on clinical data.

RESULTS

Based on simulated CGM data, the residuals generated by the particle filter are white, indicating a good model fit. For the clinical data, we use parameter values estimated based on fasting glucose data. The residuals obtained from clinical CGM data contain correlations up to lag 5.

CONCLUSION

The proposed model is shown to adequately describe the meal-induced glucose fluctuations in simulated CGM data while validations on clinical CGM data show promising results as well.

SIGNIFICANCE

The proposed model may lay the grounds for new ways of utilizing available CGM data, including CGM-based parameter estimation and stochastic optimal control.

Discrimination of simultaneous psychological and physical stressors using wristband biosignals

BACKGROUND AND OBJECTIVE

In this work, we address the problem of detecting and discriminating acute psychological stress (APS) in the presence of concurrent physical activity (PA) using wristband biosignals. We focused on signals available from wearable devices that can be worn in daily life because the ultimate objective of this work is to provide APS and PA information in real-time management of chronic conditions such as diabetes by automated personalized insulin delivery. Monitoring APS noninvasively throughout free-living conditions remains challenging because the responses to APS and PA of many physiological variables measured by wearable devices are similar.

METHODS

Various classification algorithms are compared to simultaneously detect and discriminate the PA (sedentary state, treadmill running, and stationary bike) and the type of APS (non-stress state, mental stress, and emotional anxiety). The impact of APS inducements is verified with commonly used self-reported questionnaires (The State-Trait Anxiety Inventory (STAI)). To aid the classification algorithms, novel features are generated from the physiological variables reported by a wristband device during 117 hours of experiments involving simultaneous APS inducement and PA. We also translate the APS assessment into a quantitative metric for use in predicting the adverse outcomes.

RESULTS

An accurate classification of the concurrent PA and APS states is achieved with an overall classification accuracy of 99% for PA and 92% for APS. The average accuracy of APS detection during sedentary state, treadmill running, and stationary bike is 97.3, 94.1, and 84.5%, respectively.

CONCLUSIONS

The simultaneous assessment of APS and PA throughout free-living conditions from a convenient wristband device is useful for monitoring the factors contributing to an elevated risk of acute events in people with chronic diseases like cardiovascular complications and diabetes.

Determining Physical Activity Characteristics from Wristband Data for Use in Automated Insulin Delivery Systems

Algorithms that can determine the type of physical activity (PA) and quantify the intensity can allow precision medicine approaches, such as automated insulin delivery systems that modulate insulin administration in response to PA. In this work, data from a multi-sensor wristband is used to design classifiers to distinguish among five different physical states (PS) (resting, activities of daily living, running, biking, and resistance training), and to develop models to estimate the energy expenditure (EE) of the PA for diabetes therapy. The data collected are filtered, features are extracted from the reconciled signals, and the extracted features are used by machine learning algorithms, including deep-learning techniques, to obtain accurate PS classification and EE estimation. The various machine learning techniques have different success rates ranging from 75.7% to 94.8% in classifying the five different PS. The deep neural network model with long short-term memory has 94.8% classification accuracy. We achieved 0.5 MET (Metabolic Equivalent of Task) root-mean-square error for EE estimation accuracy, relative to indirect calorimetry with randomly selected testing data (10% of collected data). We also demonstrate a 5% improvement in PS classification accuracy and a 0.34 MET decrease in the mean absolute error when using multi-sensor approach relative to using only accelerometer data.

Artificial Intelligence in Decision Support Systems for Type 1 Diabetes

Type 1 diabetes (T1D) is a chronic health condition resulting from pancreatic beta cell dysfunction and insulin depletion. While automated insulin delivery systems are now available, many people choose to manage insulin delivery manually through insulin pumps or through multiple daily injections. Frequent insulin titrations are needed to adequately manage glucose, however, provider adjustments are typically made every several months. Recent automated decision support systems incorporate artificial intelligence algorithms to deliver personalized recommendations regarding insulin doses and daily behaviors. This paper presents a comprehensive review of computational and artificial intelligence-based decision support systems to manage T1D. Articles were obtained from PubMed, IEEE Xplore, and ScienceDirect databases. No time period restrictions were imposed on the search. After removing off-topic articles and duplicates, 562 articles were left to review. Of those articles, we identified 61 articles for comprehensive review based on algorithm evaluation using real-world human data, in silico trials, or clinical studies. We grouped decision support systems into general categories of (1) those which recommend adjustments to insulin and (2) those which predict and help avoid hypoglycemia. We review the artificial intelligence methods used for each type of decision support system, and discuss the performance and potential applications of these systems.

Controlling the AP Controller: Controller Performance Assessment and Modification

BACKGROUND

Despite recent advances in closed-loop control of blood glucose concentration (BGC) in people with type 1 diabetes (T1D), online performance assessment and modification of artificial pancreas (AP) control systems remain a challenge as the metabolic characteristics of users change over time.

METHODS

A controller performance assessment and modification system (CPAMS) analyzes the glucose concentration variations and controller behavior, and modifies the parameters of the control system used in the multivariable AP system. Various indices are defined to quantitatively evaluate the controller performance in real time. Controller performance assessment and modification system also incorporates online learning from historical data to anticipate impending disturbances and proactively counteract their effects.

RESULTS

Using a multivariable simulation platform for T1D, the CPAMS is used to enhance the BGC regulation in people with T1D by means of automated insulin delivery with an adaptive learning predictive controller. Controller performance assessment and modification system increases the percentage of time in the target range (70-180) mg/dL by 52.3% without causing any hypoglycemia and hyperglycemia events.

CONCLUSIONS

The results demonstrate a significant improvement in the multivariable AP controller performance by using CPAMS.

In Silico Analysis of an Exercise-Safe Artificial Pancreas With Multistage Model Predictive Control and Insulin Safety System

BACKGROUND

Maintaining glycemic equilibrium can be challenging for people living with type 1 diabetes (T1D) as many factors (eg, length, type, duration, insulin on board, stress, and training) will impact the metabolic changes triggered by physical activity potentially leading to both hypoglycemia and hyperglycemia. Therefore, and despite the noted health benefits, many individuals with T1D do not exercise as much as their healthy peers. While technology advances have improved glucose control during and immediately after exercise, it remains one of the key limitations of artificial pancreas (AP) systems, largely because stopping insulin at the onset of exercise may not be enough to prevent impending, exercise-induced hypoglycemia.

METHODS

A hybrid AP algorithm with subject-specific exercise behavior recognition and anticipatory action is designed to prevent hypoglycemic events during and after moderate-intensity exercise. Our approach relies on a number of key innovations, namely, an activity informed premeal bolus calculator, personalized exercise pattern recognition, and a multistage model predictive control (MS-MPC) strategy that can transition between reactive and anticipatory modes. This AP design was evaluated on 100 in silico subjects from the most up-to-date FDA-accepted UVA/Padova metabolic simulator, emulating an outpatient clinical trial setting. Results with a baseline controller, a regular MPC (rMPC), are also included for comparison purposes.

RESULTS

In silico experiments reveal that the proposed MS-MPC strategy markedly reduces the number of exercise-related hypoglycemic events (8 vs 68).

CONCLUSION

An anticipatory mode for insulin administration of a monohormonal AP controller reduces the occurrence of hypoglycemia during moderate-intensity exercise.

Automated Insulin Delivery Algorithms

IN BRIEF Automated insulin delivery (AID; also known as artificial pancreas) has improved the regulation of blood glucose concentrations, reduced the frequency of hyperglycemic and hypoglycemic episodes, and improved the quality of life of people with diabetes and their families. Three different types of algorithms-proportional-integral-derivative control, model predictive control, and fuzzy-logic knowledge-based systems-have been used in AID control systems. This article will highlight the foundations of these algorithms and discuss their strengths and limitations. Multivariable artificial pancreas and dual-hormone (insulin and glucagon) systems will be introduced.

Optimal regulation of blood glucose level in Type I diabetes using insulin and glucagon

The Glucose-Insulin-Glucagon nonlinear model accurately describes how the body responds to exogenously supplied insulin and glucagon in patients affected by Type I diabetes. Based on this model, we design infusion rates of either insulin (monotherapy) or insulin and glucagon (dual therapy) that can optimally maintain the blood glucose level within desired limits after consumption of a meal and prevent the onset of both hypoglycemia and hyperglycemia. This problem is formulated as a nonlinear optimal control problem, which we solve using the numerical optimal control package PSOPT. Interestingly, in the case of monotherapy, we find the optimal solution is close to the standard method of insulin based glucose regulation, which is to assume a variable amount of insulin half an hour before each meal. We also find that the optimal dual therapy (that uses both insulin and glucagon) is better able to regulate glucose as compared to using insulin alone. We also propose an ad-hoc rule for both the dosage and the time of delivery of insulin and glucagon.