Modeling, identification and nonlinear model predictive control of type I diabetic patient

Blood Glucose Prediction from Nutrition Analytics in Type 1 Diabetes: A Review

INTRODUCTION

Type 1 Diabetes (T1D) affects over 9 million worldwide and necessitates meticulous self-management for blood glucose (BG) control. Utilizing BG prediction technology allows for increased BG control and a reduction in the diabetes burden caused by self-management requirements. This paper reviews BG prediction models in T1D, which include nutritional components.

METHOD

A systematic search, utilizing the PRISMA guidelines, identified articles focusing on BG prediction algorithms for T1D that incorporate nutritional variables. Eligible studies were screened and analyzed for model type, inclusion of additional aspects in the model, prediction horizon, patient population, inputs, and accuracy.

RESULTS

The study categorizes 138 blood glucose prediction models into data-driven (54%), physiological (14%), and hybrid (33%) types. Prediction horizons of ≤30 min are used in 36% of models, 31-60 min in 34%, 61-90 min in 11%, 91-120 min in 10%, and >120 min in 9%. Neural networks are the most used data-driven technique (47%), and simple carbohydrate intake is commonly included in models (data-driven: 72%, physiological: 52%, hybrid: 67%). Real or free-living data are predominantly used (83%).

CONCLUSION

The primary goal of blood glucose prediction in T1D is to enable informed decisions and maintain safe BG levels, considering the impact of all nutrients for meal planning and clinical relevance.

Neural network-based model predictive control for type 1 diabetic rats on artificial pancreas system

Artificial pancreas system (APS) is a viable option to treat diabetic patients. Researchers, however, have not conclusively determined the best control method for APS. Due to intra-/inter-variability of insulin absorption and action, an individualized algorithm is required to control blood glucose level (BGL) for each patient. To this end, we developed model predictive control (MPC) based on artificial neural networks (ANNs), which combines ANN for BGL prediction based on inputs and MPC for BGL control based on the ANN (NN-MPC). First, we developed a mathematical model for diabetic rats, which was used to identify individual virtual subjects by fitting to empirical data collected through an APS, including BGL data, insulin injection, and food intake. Then, the virtual subjects were used to generate datasets for training ANNs. The NN-MPC determines control actions (insulin injection) based on BGL predicted by the ANN. To evaluate the NN-MPC, we conducted experiments using four virtual subjects under three different scenarios. Overall, the NN-MPC maintained BGL within the normal range about 90% of the time with a mean absolute deviation of 4.7 mg/dl from a desired BGL. Our findings suggest that the NN-MPC can provide subject-specific BGL control in conjunction with a closed-loop APS. Graphical abstract ᅟ.

Resveratrol shows neuronal and vascular-protective effects in older, obese, streptozotocin-induced diabetic rats

Diabetes-induced CVD is the most significant complication of prolonged hyperglycaemia. The aim of this study was to determine whether resveratrol, a polyphenol antioxidant compound, when administered at a dose that can be reasonably obtained through supplementation could prevent the development of cardiovascular complications in older, obese, diabetic rats. Diabetes was induced in 6-month old, obese, male Wistar rats via a single intravenous dose of streptozotocin (65 mg/kg). Randomly selected animals were administered resveratrol (2 mg/kg) via oral gavage daily for 8 weeks. Body weights, blood glucose levels, food intake and water consumption were monitored, and assessments of vascular reactivity, tactile allodynia and left ventricular function were performed. Resveratrol therapy significantly improved tactile allodynia and vascular contractile functionality in diabetic rats (P<0·05). There were no significant changes in standardised vasorelaxation responses, plasma glucose concentrations, water consumption, body weight, left ventricular hypertrophy, kidney hypertrophy, heart rate or left ventricular compliance with resveratrol administration. Resveratrol-mediated improvements in vascular and nerve function in old, obese, diabetic rats were associated with its reported antioxidant effects. Resveratrol did not improve cardiac function nor mitigate the classic clinical symptoms of diabetes mellitus (i.e. hyperglycaemia, polydypsia and a failure to thrive). This suggests that supplementation with resveratrol at a dose achievable with commercially available supplements would not produce significant cardioprotective effects in people with diabetes mellitus.

Evaluation of a model for glycemic prediction in critically ill surgical patients

We evaluated a neural network model for prediction of glucose in critically ill trauma and post-operative cardiothoracic surgical patients. A prospective, feasibility trial evaluating a continuous glucose-monitoring device was performed. After institutional review board approval, clinical data from all consenting surgical intensive care unit patients were converted to an electronic format using novel software. This data was utilized to develop and train a neural network model for real-time prediction of serum glucose concentration implementing a prediction horizon of 75 minutes. Glycemic data from 19 patients were used to “train” the neural network model. Subsequent real-time simulated testing was performed in 5 patients to whom the neural network model was naive. Performance of the model was evaluated by calculating the mean absolute difference percent (MAD%), Clarke Error Grid Analysis, and calculation of the percent of hypoglycemic (≤70 mg/dL), normoglycemic (>70 and <150 mg/dL), and hyperglycemic (≥150 mg/dL) values accurately predicted by the model; 9,405 data points were analyzed. The models successfully predicted trends in glucose in the 5 test patients. Clark Error Grid Analysis indicated that 100.0% of predictions were clinically acceptable with 87.3% and 12.7% of predicted values falling within regions A and B of the error grid respectively. Overall model error (MAD%) was 9.0% with respect to actual continuous glucose modeling data. Our model successfully predicted 96.7% and 53.6% of the normo- and hyperglycemic values respectively. No hypoglycemic events occurred in these patients. Use of neural network models for real-time prediction of glucose in the surgical intensive care unit setting offers healthcare providers potentially useful information which could facilitate optimization of glycemic control, patient safety, and improved care. Similar models can be implemented across a wider scale of biomedical variables to offer real-time optimization, training, and adaptation that increase predictive accuracy and performance of therapies.

Development of a neural network for prediction of glucose concentration in type 1 diabetes patients

BACKGROUND

A major difficulty in the management of diabetes is the optimization of insulin therapies to avoid occurrences of hypoglycemia and hyperglycemia. Many factors impact glucose fluctuations in diabetes patients, such as insulin dosage, nutritional intake, daily activities and lifestyle (e.g., sleep-wake cycles and exercise), and emotional states (e.g., stress). The overall effect of these factors has not been fully quantified to determine the impact on subsequent glycemic trends. Recent advances in diabetes technology such as continuous glucose monitoring (CGM) provides significant sources of data, such that quantification may be possible. Depending on the CGM technology utilized, the sampling frequency ranges from 1-5 min. In this study, an intensive electronic diary documenting the factors previously described was created. This diary was utilized by 18 patients with insulin-dependent diabetes mellitus in conjunction with CGM. Utilizing this dataset, various neural network models were constructed to predict glucose in these diabetes patients while varying the predictive window from 50-180 min. The predictive capability of each neural network within the fully trained dataset was analyzed as well as the predictive capabilities of the neural networks on unseen data.

METHODS

Neural network models were created using NeuroSolutions software with variable predictive windows of 50, 75, 100, 120, 150, and 180 min. Neural network models were trained using patient datasets ranging from 11-17 patients and evaluated on patient data not included in the neural network formulation. Performance analysis was completed for the neural network models using MATLAB. Performance measures include the calculation of the mean absolute difference percent overall and at hypoglycemic and hyperglycemic extremes, and the percentage of hypoglycemic and hyperglycemic occurrences were predicted.

RESULTS

Overall, the neural network models perform adequately at predicting at normal (>70 and <180 mg/dl) and hyperglycemic ranges (> or =180 mg/dl); however, glucose concentrations in areas of hypoglycemia were commonly overestimated. One potential reason for the "high" predictions in areas of hypoglycemia is due to the minimal occurrences of hypoglycemic events within the training data. The entire 18-patient dataset (consisting of 18,400 glucose values) had a relatively low incidence of hypoglycemia (1460 CGM values < or =70 mg/dl), which corresponds to approximately 7.9% of the dataset. On the contrary, hyperglycemia comprised approximately 35.7% of the dataset (6560 CGM values >or =180 mg/dl), and euglycemic values allotted for 56.4% of the dataset (10,380 CGM values >70 and <180 mg/dl). Results further indicate that an increase in predictive window leads to a decrease in predictive accuracy of the neural network model. It is hypothesized that the underestimation of hyperglycemic extremes is due to the extension of the predictive window and the associated inability of the neural network to determine oscillations and trends in glycemia as well as the occurrence of other relevant input events such as lifestyle, emotional states, insulin dosages, and meals, which may occur within the predicted time window and may impact or change neural network weights.

CONCLUSIONS

In this investigation, the feasibility of utilizing neural network models for the prediction of glucose using predictive windows ranging from 50-180 min is demonstrated. The predictive windows were chosen arbitrarily to cover a wide range; however, longer predictive windows were implemented to gain a predictive view of 120-180 min, which is very important for diabetes patients, specifically after meals and insulin dosages. Neural networks, such as those generated in this investigation, could be utilized in a semiclosed-loop device for guiding therapy in diabetes patients. Use of such a device may lead to better glycemic control and subsequent avoidance of complications.

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.

Neural Network Analysis of Ex-vivo Expansion of Hematopoietic Stem Cells

The shortage of hematopoietic stem cells (HSCs) greatly limits their widespread clinical applications. Few studies however, investigated the relationship between the cellular expansion and the influencing factors although wide variety results of the ex-vivo expansion of HSCs existed in literature. Here, a back-propagation (BP) neural network model was employed to evaluate the ex-vivo expansions of nuclear cells (NCs), CD34(+) cells, and colony-forming units (CFU-Cs), where the output was the cellular expansion folds and the inputs include inoculated density, cytokines, resources, serum, stroma, culture time, and bioreactor types. Around 124, 86, and 90 samples were used to train the neural network for the expansion evaluations of NCs, CD34(+ )cells, and CFU-Cs, respectively, while 17, 14, and 10 samples were applied to predict respectively. The results show that for the training of network, the interval accuracy of the expansion folds for the different cells is 85.5, 86.1, and 86.7%, respectively, while the truth-value accuracy is still up to 59.7, 50.0, and 62.2%, respectively within a relative error (RE) of +/-20%. For the prediction of network, the interval accuracy can be up to 82.4, 71.4, and 70%, respectively, while the truth-value accuracy is only 29.4, 14.3, and 50.0%, respectively (RE = +/-20%). Moreover, six verification experiments were carried out based on our interval predicted values and the results proved that the five group predicted conditions lead to the correct expansion of the HSCs with the accuracy more than 80%. Considering the complexity of HSC expansion and complicated wide range of the experimental data, such relatively high interval accuracy for training and prediction as well as verification are satisfied. Therefore this nonlinear modeling makes it possible to describe quantitatively the effects of the culture conditions on the HSC expansion and to predict the optimal culture conditions for higher ex-vivo expansion of HSCs.

The Virtual NOD Mouse: Applying Predictive Biosimulation to Research in Type 1 Diabetes

Type 1 diabetes is a complex, multifactorial disease characterized by T cell-mediated autoimmune destruction of insulin-secreting pancreatic beta cells. To facilitate research in type 1 diabetes, a large-scale dynamic mathematical model of the female non-obese diabetic (NOD) mouse was developed. In this model, termed the Entelos Type 1 Diabetes PhysioLab platform, virtual NOD mice are constructed by mathematically representing components of the immune system and islet beta cell physiology important for the pathogenesis of type 1 diabetes. This report describes the scope of the platform and illustrates some of its capabilities. Specifically, using two virtual NOD mice with either average or early diabetes-onset times, we demonstrate the reproducibility of experimentally observed dynamics involved in diabetes progression, therapeutic responses to exogenous IL-10, and heterogeneity in disease onset. Additionally, we use the Type 1 Diabetes PhysioLab platform to investigate the impact of disease heterogeneity on the effectiveness of exogenous IL-10 therapy to prevent diabetes onset. Results indicate that the inability of a previously published IL-10 therapy protocol to protect NOD mice who exhibit early diabetes onset is due to high levels of pancreatic lymph node (PLN) inflammation, islet infiltration, and beta cell destruction at the time of treatment initiation. Further, simulation indicates that earlier administration of the treatment protocol can prevent NOD mice from developing diabetes by initiating treatment during the period when the disease is still sensitive to IL-10's protective function.

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.