An insulin infusion advisory system based on autotuning nonlinear model-predictive control

This paper aims at the development and evaluation of a personalized insulin infusion advisory system (IIAS), able to provide real-time estimations of the appropriate insulin infusion rate for type 1 diabetes mellitus (T1DM) patients using continuous glucose monitors and insulin pumps. The system is based on a nonlinear model-predictive controller (NMPC) that uses a personalized glucose-insulin metabolism model, consisting of two compartmental models and a recurrent neural network. The model takes as input patient's information regarding meal intake, glucose measurements, and insulin infusion rates, and provides glucose predictions. The predictions are fed to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. An algorithm based on fuzzy logic has been developed for the on-line adaptation of the NMPC control parameters. The IIAS has been in silico evaluated using an appropriate simulation environment (UVa T1DM simulator). The IIAS was able to handle various meal profiles, fasting conditions, interpatient variability, intraday variation in physiological parameters, and errors in meal amount estimations.

Diethylhexylphthalate extracted by typical newborn lipid emulsions from polyvinylchloride infusion systems causes significant changes in histology of rabbit

Background: Looking for a candidate substance inducing hepatobiliary dysfunction under parenteral nutrition (PN) in newborns, we recently discovered that newborn infusions extract large amounts of the plasticizer diethylhexylphthalate (DEHP) from commonly used polyvinylchloride (PVC) infusion lines. This plasticizer is well known to be genotoxic and teratogenic in animals and to cause changes in various organs and enzyme systems even in humans. The aim of this study was to examine the effect of DEHP, extracted in the same way and in the same amount as in newborns, on livers of young rabbits. Methods: Prepubertal rabbits received lipid emulsion through central IV lines continuously for 3 weeks either via PVC or polyethylene (PE) infusion systems. Livers were examined after 1 and 3 weeks by light and electron microscopy. Results: By light microscopy, hydropic degeneration, single-cell necrosis, fibrosis, and bile duct proliferation were observed more in the PVC group. Electron microscopy revealed multiple nuclear changes, clusters and atypical forms of peroxisomes, proliferation of smooth endoplasmic reticulum, increased deposition of lipofuscin, and a mild perisinusoidal fibrosis only in the PVC group. These changes, which are generally regarded as reaction upon a toxic stimulus, could be exclusively attributed to DEHP. Conclusions: This investigation proved that DEHP produces toxin-like changes in livers of young rabbits in the same dose, duration, and method of administration as in newborn infants. For this reason, it is likely that DEHP is the substance that causes hepatobiliary dysfunction in newborns under PN. Possible modes of action of DEHP are proposed.

Continuous subcutaneous insulin infusion therapy: effects on quality of life

To compare the diabetes-specific quality of life in subjects with type 1 diabetes treating their diabetes with multiple daily injections (MDI) to that of subjects on continuous subcutaneous insulin infusion (CSII).

Establishing glycaemic control with continuous subcutaneous insulin infusion in children and adolescents with type 1 diabetes: experience of the PedPump Study in 17 countries

AIMS/HYPOTHESIS: To assess the use of paediatric continuous subcutaneous infusion (CSII) under real-life conditions by analysing data recorded for up to 90 days and relating them to outcome. METHODS: Pump programming data from patients aged 0-18 years treated with CSII in 30 centres from 16 European countries and Israel were recorded during routine clinical visits. HbA(1c) was measured centrally. RESULTS: A total of 1,041 patients (age: 11.8 +/- 4.2 years; diabetes duration: 6.0 +/- 3.6 years; average CSII duration: 2.0 +/- 1.3 years; HbA(1c): 8.0 +/- 1.3% [means +/- SD]) participated. Glycaemic control was better in preschool (n = 142; 7.5 +/- 0.9%) and pre-adolescent (6-11 years, n = 321; 7.7 +/- 1.0%) children than in adolescent patients (12-18 years, n = 578; 8.3 +/- 1.4%). There was a significant negative correlation between HbA(1c) and daily bolus number, but not between HbA(1c) and total daily insulin dose. The use of <6.7 daily boluses was a significant predictor of an HbA(1c) level >7.5%. The incidence of severe hypoglycaemia and ketoacidosis was 6.63 and 6.26 events per 100 patient-years, respectively. CONCLUSIONS/INTERPRETATION: This large paediatric survey of CSII shows that glycaemic targets can be frequently achieved, particularly in young children, and the incidence of acute complications is low. Adequate substitution of basal and prandial insulin is associated with a better HbA(1c).

A reduction in severe hypoglycaemia in type 1 diabetes in a randomized crossover study of continuous intraperitoneal compared with subcutaneous insulin infusion

Continuous intraperitoneal insulin infusion (CIPII) with the DiaPort system using regular insulin was compared to continuous subcutaneous insulin infusion (CSII) using insulin Lispro, to investigate the frequency of hypoglycemia, blood glucose control, quality of life, and safety.

An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.

Effects of colostrum feeding and glucocorticoid administration on insulin-dependent glucose metabolism in neonatal calves

Colostrum feeding and glucocorticoid administration affect glucose metabolism and insulin release in calves. We have tested the hypothesis that dexamethasone as well as colostrum feeding influence insulin-dependent glucose metabolism in neonatal calves using the euglycemic-hyperinsulinemic clamp technique. Newborn calves were fed either colostrum or a milk-based formula (n=14 per group) and in each feeding group, half of the calves were treated with dexamethasone (30 microg/[kg body weight per day]). Preprandial blood samples were taken on days 1, 2, and 4. On day 5, insulin was infused for 3h and plasma glucose concentrations were kept at 5 mmol/L+/-10%. Clamps were combined with [(13)C]-bicarbonate and [6,6-(2)H]-glucose infusions for 5.5h (i.e., from -150 to 180 min, relative to insulin infusion) to determine glucose turnover, glucose appearance rate (Ra), endogenous glucose production (eGP), and gluconeogenesis before and at the end of the clamp. After the clamp liver biopsies were taken to measure mRNA levels of phosphoenolpyruvate carboxykinase (PEPCK) and pyruvate carboxylase (PC). Dexamethasone increased plasma glucose, insulin, and glucagon concentrations in the pre-clamp period thus necessitating a reduction in the rate of glucose infusion to maintain euglycemia during the clamp. Glucose turnover and Ra increased during the clamp and were lower at the end of the clamp in dexamethasone-treated calves. Dexamethasone treatment did not affect basal gluconeogenesis or eGP. At the end of the clamp, dexamethasone reduced eGP and PC mRNA levels, whereas mitochondrial PEPCK mRNA levels increased. In conclusion, insulin increased glucose turnover and dexamethasone impaired insulin-dependent glucose metabolism, and this was independent of different feeding.

Conjoint regulation of glucagon concentrations via plasma insulin and glucose in dairy cows

Insulin and glucagon are glucoregulatory hormones that contribute to glucose homeostasis. Plasma insulin is elevated during normoglycemia or hyperglycemia and acts as a suppressor of glucagon secretion. We have investigated if and how insulin and glucose contribute to the regulation of glucagon secretion through long term (48 h) elevated insulin concentrations during simultaneous hypoglycemia or euglycemia in mid-lactating dairy cows. Nineteen Holstein dairy cows were randomly assigned to 3 treatment groups: an intravenous insulin infusion (HypoG, n = 5) to decrease plasma glucose concentrations (2.5 mmol/L), a hyperinsulinemic-euglycemic clamp to study effects of insulin at simultaneously normal glucose concentrations (EuG, n = 6) and a 0.9% saline infusion (NaCl, n = 8). Plasma glucose was measured at 5-min intervals, and insulin and glucose infusion rates were adjusted accordingly. Area under the curve of hourly glucose, insulin, and glucagon concentrations on day 2 of infusion was evaluated by analysis of variance with treatments as fixed effect. Insulin infusion caused an increase of plasma insulin area under the curve (AUC)/h in HypoG (41.9 ± 8.1 mU/L) and EuG (57.8 ± 7.8 mU/L) compared with NaCl (13.9 ± 1.1 mU/L; P < 0.01). Induced hyperinsulinemia caused a decline of plasma glucose AUC/h to 2.3 ± 0.1 mmol/L in HypoG (P < 0.01), whereas plasma glucose AUC/h remained unchanged in EuG (3.8 ± 0.2 mmol/L) and NaCl (4.1 ± 0.1 mmol/L). Plasma glucagon AUC/h was lower in EuG (84.0 ± 6.3 pg/mL; P < 0.05) and elevated in HypoG (129.0 ± 7.0 pg/mL; P < 0.01) as compared with NaCl (106.1 ± 5.4 pg/mL). The results show that intravenous insulin infusion induces elevated glucagon concentrations during hypoglycemia, although the same insulin infusion reduces glucagon concentrations at simultaneously normal glucose concentrations. Thus, insulin does not generally have an inhibitory effect on glucagon concentrations. If simultaneously glucose is low and insulin is high, glucagon is upregulated to increase glucose availability. Therefore, insulin and glucose are conjoint regulatory factors of glucagon concentrations in dairy cows, and the plasma glucose status is the key factor to decide if its concentrations are increased or decreased. This regulatory effect can be important for the maintenance of glucose homeostasis if insulin secretion is upregulated by other factors than high glucose such as high plasma lipid and protein concentrations at simultaneously low glucose.

SMARTDIAB: a communication and information technology approach for the intelligent monitoring, management and follow-up of type 1 diabetes patients

SMARTDIAB is a platform designed to support the monitoring, management, and treatment of patients with type 1 diabetes mellitus (T1DM), by combining state-of-the-art approaches in the fields of database (DB) technologies, communications, simulation algorithms, and data mining. SMARTDIAB consists mainly of two units: 1) the patient unit (PU); and 2) the patient management unit (PMU), which communicate with each other for data exchange. The PMU can be accessed by the PU through the internet using devices, such as PCs/laptops with direct internet access or mobile phones via a Wi-Fi/General Packet Radio Service access network. The PU consists of an insulin pump for subcutaneous insulin infusion to the patient and a continuous glucose measurement system. The aforementioned devices running a user-friendly application gather patient's related information and transmit it to the PMU. The PMU consists of a diabetes data management system (DDMS), a decision support system (DSS) that provides risk assessment for long-term diabetes complications, and an insulin infusion advisory system (IIAS), which reside on a Web server. The DDMS can be accessed from both medical personnel and patients, with appropriate security access rights and front-end interfaces. The DDMS, apart from being used for data storage/retrieval, provides also advanced tools for the intelligent processing of the patient's data, supporting the physician in decision making, regarding the patient's treatment. The IIAS is used to close the loop between the insulin pump and the continuous glucose monitoring system, by providing the pump with the appropriate insulin infusion rate in order to keep the patient's glucose levels within predefined limits. The pilot version of the SMARTDIAB has already been implemented, while the platform's evaluation in clinical environment is being in progress.

An Actor-Critic based controller for glucose regulation in type 1 diabetes

A novel adaptive approach for glucose control in individuals with type 1 diabetes under sensor-augmented pump therapy is proposed. The controller, is based on Actor-Critic (AC) learning and is inspired by the principles of reinforcement learning and optimal control theory. The main characteristics of the proposed controller are (i) simultaneous adjustment of both the insulin basal rate and the bolus dose, (ii) initialization based on clinical procedures, and (iii) real-time personalization. The effectiveness of the proposed algorithm in terms of glycemic control has been investigated in silico in adults, adolescents and children under open-loop and closed-loop approaches, using announced meals with uncertainties in the order of ±25% in the estimation of carbohydrates. The results show that glucose regulation is efficient in all three groups of patients, even with uncertainties in the level of carbohydrates in the meal. The percentages in the A+B zones of the Control Variability Grid Analysis (CVGA) were 100% for adults, and 93% for both adolescents and children. The AC based controller seems to be a promising approach for the automatic adjustment of insulin infusion in order to improve glycemic control. After optimization of the algorithm, the controller will be tested in a clinical trial.

Real-time adaptive models for the personalized prediction of glycemic profile in type 1 diabetes patients

Prediction of glycemic profile is an important task for both early recognition of hypoglycemia and enhancement of the control algorithms for optimization of insulin infusion rate. Adaptive models for glucose prediction and recognition of hypoglycemia based on statistical and artificial intelligence techniques are presented.

Anesthesia or sedation for gastroenterologic endoscopies

PURPOSE OF REVIEW: Because propofol is the sedative preferred by gastroenterologists, we focus this review on gastroenterologist-directed propofol sedation, provide simulations of the respiratory depressant effect of different dosing protocols and give a perspective on future developments in computer-assisted sedation techniques. RECENT FINDINGS: Propofol use by nonanesthesiologists remains a contraindication in the package insert of propofol in most countries. Sedation guidelines produced by the American Society of Gastroenterology partially contradict those produced by the American Society of Anesthesiologists for sedation by nonanesthesiologists, whereas the German guidelines were developed with anesthesiologists involved. The use of fospropofol, recently approved by the US Food and Drug Administration for sedation, is considered an alternative to propofol by some gastroenterologists. Methodological errors in earlier pharmacological studies have to be solved before widespread use of fospropofol is justified, however. Our simulations show that dosing protocols with small boluses administered at reasonable intervals induce less respiratory depression than large boluses. Interindividual variability of propofol-induced respiratory depression is illustrated by different pharmacokinetic and dynamic parameter sets used in the simulation. Two computer-assisted propofol infusion systems are currently being investigated. They not only incorporate the target effect but also the side effects, which may limit respiratory depression. SUMMARY: Propofol use by gastroenterologists may be well tolerated if appropriate patient selection, staff training, monitoring and low-dose sedation protocols are applied.

Personalized Tuning of a Reinforcement Learning Control Algorithm for Glucose Regulation

Artificial pancreas is in the forefront of research towards the automatic insulin infusion for patients with type 1 diabetes. Due to the high inter- and intra-variability of the diabetic population, the need for personalized approaches has been raised. This study presents an adaptive, patient-specific control strategy for glucose regulation based on reinforcement learning and more specifically on the Actor-Critic (AC) learning approach. The control algorithm provides daily updates of the basal rate and insulin-to-carbohydrate (IC) ratio in order to optimize glucose regulation. A method for the automatic and personalized initialization of the control algorithm is designed based on the estimation of the transfer entropy (TE) between insulin and glucose signals. The algorithm has been evaluated in silico in adults, adolescents and children for 10 days. Three scenarios of initialization to i) zero values, ii) random values and iii) TE-based values have been comparatively assessed. The results have shown that when the TE-based initialization is used, the algorithm achieves faster learning with 98%, 90% and 73% in the A+B zones of the Control Variability Grid Analysis for adults, adolescents and children respectively after five days compared to 95%, 78%, 41% for random initialization and 93%, 88%, 41% for zero initial values. Furthermore, in the case of children, the daily Low Blood Glucose Index reduces much faster when the TE-based tuning is applied. The results imply that automatic and personalized tuning based on TE reduces the learning period and improves the overall performance of the AC algorithm.

GoCARB: Smartphone-App zur Kohlenhydratberechnung bei Typ-1-Diabetes

Smartphone-App zur Kohlenhydratberechnung Neue Technologien wie Blutzuckersensoren und moderne Insulinpumpen prägten die Therapie des Typ-1-Diabetes (T1D) in den letzten Jahren in wesentlichem Ausmaß. Smartphones sind aufgrund ihrer rasanten technischen Entwicklung eine weitere Plattform für Applikationen zur Therapieunterstützung bei T1D. GoCARB Hierbei handelt es sich um ein zur Kohlenhydratberechnung entwickeltes System für Personen mit T1D. Die Basis für Endanwender stellt ein Smartphone mit Kamera dar. Zur Berechnung werden 2 mit dem Smartphone aus verschiedenen Winkeln aufgenommene Fotografien einer auf einem Teller angerichteten Mahlzeit benötigt. Zusätzlich ist eine neben dem Teller platzierte Referenzkarte erforderlich. Die Grundlage für die Kohlenhydratberechnung ist ein Computer-Vision-gestütztes Programm, das die Mahlzeiten aufgrund ihrer Farbe und Textur erkennt. Das Volumen der Mahlzeit wird mit Hilfe eines dreidimensional errechneten Modells bestimmt. Durch das Erkennen der Art der Mahlzeiten sowie deren Volumen kann GoCARB den Kohlenhydratanteil unter Einbeziehung von Nährwerttabellen berechnen. Für die Entwicklung des Systems wurde eine Bilddatenbank von mehr als 5000 Mahlzeiten erstellt und genutzt. Resümee Das GoCARB-System befindet sich aktuell in klinischer Evaluierung und ist noch nicht für Patienten verfügbar.

Effects of an overnight intravenous lipid infusion on intramyocellular lipid content and insulin sensitivity in African-American versus Caucasian adolescents

To explain the predisposition for insulin resistance among African American (AA) adolescents, this study aimed to: 1) examine changes in intramyocellular lipid content (IMCL), and insulin sensitivity with intralipid (IL) infusion; and 2) determine whether the increase in IMCL is comparable between AA and Caucasian adolescents.