This work presents research in machine learning for diabetes management. There are two major contributions: (1) development of a metric for measuring glycemic variability, a serious problem for patients with diabetes; and (2) predicting patient blood glucose levels, in order to preemptively detect and avoid potential health problems. The glycemic variability metric uses machine learning trained on multiple statistical and domain specific features to match physician consensus of glycemic variability. The metric performs similarly to an individual physician's ability to match the consensus. When used as a screen for detecting excessive glycemic variability, the metric outperforms the baseline metrics. The blood glucose prediction model uses machine learning to integrate a general physiological model and life-events to make patient-specific predictions 30 and 60 minutes in the future. The blood glucose prediction model was evaluated in several situations such as near a meal or during exercise. The prediction model outperformed the baselines prediction models, and performed similarly to, and in some cases outperformed, expert physicians who were given the same prediction problems.

This book tackles the problem of overshoot and undershoot in blood glucose levels caused by delay in the effects of carbohydrate consumption and insulin administration. The ideas presented here will be very important in maintaining the welfare of insulin-dependent diabetics and avoiding the damaging effects of unpredicted swings in blood glucose – accurate prediction enables the implementation of counter-measures. The glucose prediction algorithms described are also a key and critical ingredient of automated insulin delivery systems, the so-called “artificial pancreas”. The authors address the topic of blood-glucose prediction from medical, scientific and technological points of view. Simulation studies are utilized for complementary analysis but the primary focus of this book is on real applications, using clinical data from diabetic subjects. The text details the current state of the art by surveying prediction algorithms, and then moves beyond it with the most recent advances in data-based modeling of glucose metabolism. The topic of performance evaluation is discussed and the relationship of clinical and technological needs and goals examined with regard to their implications for medical devices employing prediction algorithms. Practical and theoretical questions associated with such devices and their solutions are highlighted. This book shows researchers interested in biomedical device technology and control researchers working with predictive algorithms how incorporation of predictive algorithms into the next generation of portable glucose measurement can make treatment of diabetes safer and more efficient.

Intended for diabetes researchers and medical professionals who work closely with patients with diabetes, this newly updated and expanded edition provides new perspectives and direct insight into the causes and consequences of this serious medical condition from one of the foremost experts in the field. Using the latest scientific and medical developments and trends, readers will learn how to identify, prevent, and treat this challenging phenomenon within the parameters of the diabetes care regimen.

This book provides comprehensive information on continuous glucose monitoring (CGM). The first section focuses on the fundamentals of CGM technology, including the principles of CGM, accuracy assessment, operation procedure, management processes, the picture-interpretation methodology, the clinical value of CGM parameters, reference values, clinical applications of CGM report and management systems, and clinical indications. In turn, the second section describes the clinical application of CGM, including assessing blood glucose fluctuation and hypoglycemic effects, detecting hypoglycemia and identifying fasting hyperglycemia. It also describes the role of CGM in connection with specific diseases, such as fulminant type 1 diabetes, gestational diabetes mellitus, steroid diabetes, and insulinoma. The closing chapter outlines the future of CGM. In addition, the book presents typical cases and analyses of nearly a hundred typical monitoring maps. As such, it offers diabetic health care doctors a valuable reference guide to the clinical application of and scientific research on CGM.

Critical care (e.g. trauma and cardiothoracic surgical) and diabetic patients are prone to variability in glucose concentration on a daily basis. Hypoglycemic and hyperglycemic glucose values in these patient populations have been associated with decreased patient outcomes. In diabetic patients, persistently elevated glucose values are associated with development of long term complications such as, but not limited to retinopathy, neuropathy, and nephropathy. In the critical care patient population, elevated glucose has been correlated to increases in mortality, length of stay in the intensive care unit (ICU), and morbidities. The maintenance of tight glycemic control in these patients without severe hypoglycemia or glycemic variability appears to improve outcomes in these patients. Various factors are associated with future glycemic excursions such as, but not limited to: lifestyle/activities (e.g. sleep-wake cycles), emotional factors (e.g. stress), nutritional intake, medication dosages, and ICU medical records (in critical care patients). In the field of diabetes research, models for prediction of glucose and/or models used to maintain tight glycemic control have been the focus of research. In the critical care patient population, very little research into development of such models has been completed to date. Multiple factors affect or are indicators of future glucose concentration. A suitable modeling technique needs to incorporate the effect of such factors for accurate prediction of glucose. A modeling technique well suited for this task is a neural network model. A neural network is an adapative modeling technique, which learns and updates model parameters based on determining patterns/trends existent in input data. This adapative capability, makes neural network modeling well suited for prediction of glucose where multiple factors impact future glycemic excursions. This dissertation summarizes the development and optimization of various neural network model architectures for the real-time prediction of glucose in diabetic and critical care patients. Neural network models were configured to predict glucose using prediction horizons>60 minutes, which have not been attained in many predictive models to date. The performance of the neural network model is assessed via determination of overall model error, percentage of glycemic extremes predicted, and clinical acceptability of model predictions as determined via Clarke Error Grid Analysis.

As the number of patients with diabetes increases annually, it is not surprising that the number of patients with diabetes who are admitted to the hospital also increases. Once in the hospital, patients with diabetes or hyperglycemia may be admitted to the Intensive Care Unit, require urgent or elective surgery, enteral or parenteral nutrition, intravenous insulin infusion, or therapies that significantly impact glycemic control (e.g., steroids). Because many clinical outcomes are influenced by the degree of glycemic control, knowledge of the best practices in inpatient diabetes management is extremely important. The field of inpatient management of diabetes and hyperglycemia has grown substantially in the last several years. This body of knowledge is summarized in this book, so it can reach the audience of hospitalists, endocrinologists, nurses and other team members who take care of hospitalized patients with diabetes and hyperglycemia.