Reduction of number and duration of hypoglycemic events by glucose prediction methods: a proof-of-concept in silico study

Hypoglycaemia detection and prediction techniques: A systematic review on the latest developments

The main objective of diabetes control is to correct hyperglycaemia while avoiding hypoglycaemia, especially in insulin-treated patients. Fear of hypoglycaemia is a hurdle to effective correction of hyperglycaemia because it promotes under-dosing of insulin. Strategies to minimise hypoglycaemia include education and training for improved hypoglycaemia awareness and the development of technologies to allow their early detection and thus minimise their occurrence. Patients with impaired hypoglycaemia awareness would benefit the most from these technologies. The purpose of this systematic review is to review currently available or in-development technologies that support detection of hypoglycaemia or hypoglycaemia risk, and identify gaps in the research. Nanomaterial use in sensors is a promising strategy to increase the accuracy of continuous glucose monitoring devices for low glucose values. Hypoglycaemia is associated with changes on vital signs, so electrocardiogram and encephalogram could also be used to detect hypoglycaemia. Accuracy improvements through multivariable measures can make already marketed galvanic skin response devices a good noninvasive alternative. Breath volatile organic compounds can be detected by dogs and devices and alert patients at hypoglycaemia onset, while near-infrared spectroscopy can also be used as a hypoglycaemia alarms. Finally, one of the main directions of research are deep learning algorithms to analyse continuous glucose monitoring data and provide earlier and more accurate prediction of hypoglycaemia. Current developments for early identification of hypoglycaemia risk combine improvements of available 'needle-type' enzymatic glucose sensors and noninvasive alternatives. Patient usability will be essential to demonstrate to allow their implementation for daily use in diabetes management.

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.

Real-World Hypoglycemia Avoidance with a Continuous Glucose Monitoring System's Predictive Low Glucose Alert

BACKGROUND

Programmable and fixed auditory and/or vibratory threshold alerts are essential features of real-time continuous glucose monitoring (rtCGM) systems that provide users time to intervene before the onset of clinical hypoglycemia or hyperglycemia. A sixth-generation rtCGM system from Dexcom, Inc. (G6) includes a new alert that is triggered when an algorithm predicts that an estimated glucose value ≤55 mg/dL will occur within 20 min, allowing users more time to act to avoid hypoglycemia. We examined whether this predictive low glucose alert provided added benefit to traditional low threshold alerts.

METHODS

We analyzed glucose values from an anonymized sample of 1424 patients who transitioned to G6 from the preceding fifth-generation system (G5) with no predictive alert. Users with the low threshold alert setting of 70 or 80 mg/dL were evaluated separately. Receiver users, those who disabled the predictive low glucose alert, or those with <30 days of data immediately before or after the transition to G6 were excluded.

RESULTS

Percent time <54, ≤55, <70, and >250 mg/dL fell significantly after the transition to G6, independent of low threshold alert setting. Time in range improved for G6 users with a low threshold alert setting of 70 mg/dL.

CONCLUSIONS

Advance warning provided by predictive low glucose alerts may further reduce hypoglycemia among rtCGM-experienced users.

Using LSTMs to learn physiological models of blood glucose behavior

For people with type 1 diabetes, good blood glucose control is essential to keeping serious disease complications at bay. This entails carefully monitoring blood glucose levels and taking corrective steps whenever they are too high or too low. If blood glucose levels could be accurately predicted, patients could take proactive steps to prevent blood glucose excursions from occurring. However, accurate predictions require complex physiological models of blood glucose behavior. Factors such as insulin boluses, carbohydrate intake, and exercise influence blood glucose in ways that are difficult to capture through manually engineered equations. In this paper, we describe a recursive neural network (RNN) approach that uses long short-term memory (LSTM) units to learn a physiological model of blood glucose. When trained on raw data from real patients, the LSTM networks (LSTMs) obtain results that are competitive with a previous state-of-the-art model based on manually engineered physiological equations. The RNN approach can incorporate arbitrary physiological parameters without the need for sophisticated manual engineering, thus holding the promise of further improvements in prediction accuracy.

Prediction of Adverse Glycemic Events From Continuous Glucose Monitoring Signal

The most important objective of any diabetes therapy is to maintain the blood glucose concentration within the euglycemic range, avoiding or at least mitigating critical hypo/hyperglycemic episodes. Modern continuous glucose monitoring (CGM) devices bear the promise of providing the patients with an increased and timely awareness of glycemic conditions as these get dangerously near to hypo/hyperglycemia. The challenge is to detect, with reasonable advance, the patterns leading to risky situations, allowing the patient to make therapeutic decisions on the basis of future (predicted) glucose concentration levels. We underline that a technically sound performance comparison of the approaches proposed in recent years has yet to be done, thus it is unclear which one is preferred. The aim of this study is to fill this gap by carrying out a comparative analysis among the most common methods for glucose event prediction. Both regression and classification algorithms have been implemented and analyzed, including static and dynamic training approaches. The dataset consists of 89 CGM time series measured in diabetic subjects for 7 subsequent days. Performance metrics, specifically defined to assess and compare the event-prediction capabilities of the methods, have been introduced and analyzed. Our numerical results show that a static training approach exhibits better performance, in particular when regression methods are considered. However, classifiers show some improvement when trained for a specific event category, such as hyperglycemia, achieving performance comparable to the regressors, with the advantage of predicting the events sooner.

Exploring the Frequency Domain of Continuous Glucose Monitoring Signals to Improve Characterization of Glucose Variability and of Diabetic Profiles

BACKGROUND

Continuous glucose monitoring (CGM) devices measure interstitial glucose concentrations (normally every 5 minutes), allowing observation of glucose variability (GV) patterns during the whole day. This information could be used to improve prescription of treatments and of insulin dosages for people suffering diabetes. Previous efforts have been focused on proposing indices of GV either in time or glucose domains, while the frequency domain has been explored only partially. The aim of this work is to explore the CGM signal in the frequency domain to understand if new indexes or features could be identified and contribute to a better characterization of glucose variability.

METHODS

The direct fast Fourier transform (FFT) and the Welch method were used to analyze CGM signals from three different profiles: people at risk of developing type 2 diabetes (P@R), T2D patients, and type 1 diabetes (T1D) patients.

RESULTS

The results suggests that features extracted from the FFT (ie, the localization and power of the maximum peak of the power spectrum and the bandwidth at 3 dB) are able to provide a characterization for all the three populations under study compared with the Welch approach.

CONCLUSIONS

Such preliminary results can represent a good insight for futures investigations with the possibility of building and using new indexes of glucose variability based on the frequency features.

"Let the Algorithm Do the Work": Reduction of Hypoglycemia Using Sensor-Augmented Pump Therapy with Predictive Insulin Suspension (SmartGuard) in Pediatric Type 1 Diabetes Patients

BACKGROUND

A sensor-augmented insulin pump (SAP) using the MiniMed^® 640G system with SmartGuard™ technology allows an automatic stop of insulin delivery based on prediction of low glucose levels. Since pediatric patients are particularly prone to hypoglycemia, this device may offer additional protection beyond conventional sensor-augmented therapy.

METHODS

This prospective, pediatric multicenter user evaluation assessed 6 weeks of SAP with SmartGuard (threshold setting for hypoglycemia: 70 mg/dL) compared to a preceding period of 2 weeks with SAP only. The primary outcome was the potential reduction in the frequency of hypoglycemic episodes and hypoglycemic intensity (area under the curve [AUC] and time <70 mg/dL).

RESULTS

The study included 24 patients with at least 3 months of insulin pump use (average age: 11.6 ± 5.1 years, 15 female, average type 1 diabetes duration: 7.5 ± 4.2 years, mean ± SD) who had on average 3.2 ± 1.0 predictive suspensions/patient/day. The mean sensor glucose minimum during suspension was 78 ± 6 mg/dL and the average suspension time was 155 ± 47 min/day. Use of SmartGuard in patients treated as per the protocol (n = 18) reduced the number of instances in which the glucose level was <70 mg/dL (1.02 ± 0.52 to 0.72 ± 0.36; P = 0.027), as well as AUC <70 mg/dL (0.76 ± 0.73 to 0.38 ± 0.24; P = 0.027) and the time/day the level fell below 70 mg/dL (73 ± 56 to 31 ± 22 min). The reduction of hypoglycemia was not associated with a significant change in mean glucose concentration (171 ± 26 to 180 ± 19 mg/dL, P = 0.111) and HbA1c (7.5% ± 0.5% to 7.6% ± 0.7%, (P = 0.329). Manual resumption of insulin delivery followed by carbohydrate intake resulted in significantly higher glucose levels 1 h after suspension compared to SmartGuard suspensions with automatic resume (190.8 ± 26.5 vs. 138.7 ± 10.3 mg/dL; P < 0.001).

CONCLUSIONS

SmartGuard technology significantly reduced the risk for hypoglycemia in pediatric type 1 diabetes patients without increasing HbA1c. Patients must be educated that when using combining predictive low-glucose insulin suspension technology, extra carbohydrate intake in response to an alarm combined with manual resumption is likely to cause rebound hyperglycemia. The best results were achieved when the user did not interfere with pump operation.

Online prediction of glucose concentration in type 1 diabetes using extreme learning machines

We propose an online machine-learning solution to the problem of nonlinear glucose time series prediction in type 1 diabetes. Recently, extreme learning machine (ELM) has been proposed for training single hidden layer feed-forward neural networks. The high accuracy and fast learning speed of ELM drive us to investigate its applicability to the glucose prediction problem. Given that diabetes self-monitoring data are received sequentially, we focus on online sequential ELM (OS-ELM) and online sequential ELM kernels (KOS-ELM). A multivariate feature set is utilized concerning subcutaneous glucose, insulin therapy, carbohydrates intake and physical activity. The dataset comes from the continuous multi-day recordings of 15 type 1 patients in free-living conditions. Assuming stationarity and evaluating the performance of the proposed method by 10-fold cross- validation, KOS-ELM were found to perform better than OS-ELM in terms of prediction error, temporal gain and regularity of predictions for a 30-min prediction horizon.

How Much Is Short-Term Glucose Prediction in Type 1 Diabetes Improved by Adding Insulin Delivery and Meal Content Information to CGM Data? A Proof-of-Concept Study

BACKGROUND

In type 1 diabetes (T1D) management, short-term glucose prediction can allow to anticipate therapeutic decisions when hypo/hyperglycemia is imminent. Literature prediction methods mainly use past continuous glucose monitoring (CGM) readings. Sophisticated algorithms can use information on insulin delivered and meal carbohydrate (CHO) content. The quantification of how much insulin and CHO information improves glucose prediction is missing in the literature and is investigated, in an open-loop setting, in this proof-of-concept study.

METHODS

We adopted a versatile literature prediction methodology able to utilize a variety of inputs. We compared predictors that use (1) CGM; (2) CGM and insulin; (3) CGM and CHO; and (4) CGM, insulin, and CHO. Data of 15 T1D subjects in open-loop setup were used. Prediction was evaluated via absolute error and temporal gain focusing on meal/night periods. The relative importance of each individual input of the predictor was evaluated with a sensitivity analysis.

RESULTS

For a prediction horizon (PH) ≥ 30 minutes, insulin and CHO information improves prediction accuracy of 10% and double the temporal gain during the 2 hours following the meal. During the night the 4 methods did not give statistically different results. When PH ≥ 45 minutes, the influence of CHO information on prediction is 5-fold that of insulin.

CONCLUSIONS

In an open-loop setting, with PH ≥ 30 minutes, information on CHO and insulin improves short-term glucose prediction in the 2-hour time window following a meal, but not during the night. CHO information improves prediction significantly more than insulin.

Data Mining of a Remote Behavioral Tracking System for Type 2 Diabetes Patients: A Prospective Cohort Study

Background

Complications from type 2 diabetes mellitus can be prevented when patients perform health behaviors such as vigorous exercise and glucose-regulated diet. The use of smartphones for tracking such behaviors has demonstrated success in type 2 diabetes management while generating repositories of analyzable digital data, which, when better understood, may help improve care. Data mining methods were used in this study to better understand self-monitoring patterns using smartphone tracking software.

Objective

Associations were evaluated between the smartphone monitoring of health behaviors and HbA1c reductions in a patient subsample with type 2 diabetes who demonstrated clinically significant benefits after participation in a randomized controlled trial.

Methods

A priori association-rule algorithms, implemented in the C language, were applied to app-discretized use data involving three primary health behavior trackers (exercise, diet, and glucose monitoring) from 29 participants who achieved clinically significant HbA1c reductions. Use was evaluated in relation to improved HbA1c outcomes.

Results

Analyses indicated that nearly a third (9/29, 31%) of participants used a single tracker, half (14/29, 48%) used two primary trackers, and the remainder (6/29, 21%) of the participants used three primary trackers. Decreases in HbA1c were observed across all groups (0.97-1.95%), but clinically significant reductions were more likely with use of one or two trackers rather than use of three trackers (OR 0.18, P=.04).

Conclusions

Data mining techniques can reveal relevant coherent behavior patterns useful in guiding future intervention structure. It appears that focusing on using one or two trackers, in a symbolic function, was more effective (in this sample) than regular use of all three trackers.

Hypo- and Hyperglycemic Alarms: Devices and Algorithms

Soon after the discovery that insulin regulates blood glucose by Banting and Best in 1922, the symptoms and risks associated with hypoglycemia became widely recognized. This article reviews devices to warn individuals of impending hypo- and hyperglycemia; biosignals used by these devices include electroencephalography, electrocardiography, skin galvanic resistance, diabetes alert dogs, and continuous glucose monitors (CGMs). While systems based on other technology are increasing in performance and decreasing in size, CGM technology remains the best method for both reactive and predictive alarming of hypo- or hyperglycemia.

Evaluation of short-term predictors of glucose concentration in type 1 diabetes combining feature ranking with regression models

Glucose concentration in type 1 diabetes is a function of biological and environmental factors which present high inter-patient variability. The objective of this study is to evaluate a number of features, which are extracted from medical and lifestyle self-monitoring data, with respect to their ability to predict the short-term subcutaneous (s.c.) glucose concentration of an individual. Random forests (RF) and RReliefF algorithms are first employed to rank the candidate feature set. Then, a forward selection procedure follows to build a glucose predictive model, where features are sequentially added to it in decreasing order of importance. Predictions are performed using support vector regression or Gaussian processes. The proposed method is validated on a dataset of 15 type diabetics in real-life conditions. The s.c. glucose profile along with time of the day and plasma insulin concentration are systematically highly ranked, while the effect of food intake and physical activity varies considerably among patients. Moreover, the average prediction error converges in less than d/2 iterations (d is the number of features). Our results suggest that RF and RReliefF can find the most informative features and can be successfully used to customize the input of glucose models.

Model of glucose sensor error components: identification and assessment for new Dexcom G4 generation devices

It is clinically well-established that minimally invasive subcutaneous continuous glucose monitoring (CGM) sensors can significantly improve diabetes treatment. However, CGM readings are still not as reliable as those provided by standard fingerprick blood glucose (BG) meters. In addition to unavoidable random measurement noise, other components of sensor error are distortions due to the blood-to-interstitial glucose kinetics and systematic under-/overestimations associated with the sensor calibration process. A quantitative assessment of these components, and the ability to simulate them with precision, is of paramount importance in the design of CGM-based applications, e.g., the artificial pancreas (AP), and in their in silico testing. In the present paper, we identify and assess a model of sensor error of for two sensors, i.e., the G4 Platinum (G4P) and the advanced G4 for artificial pancreas studies (G4AP), both belonging to the recently presented "fourth" generation of Dexcom CGM sensors but different in their data processing. Results are also compared with those obtained by a sensor belonging to the previous, "third," generation by the same manufacturer, the SEVEN Plus (7P). For each sensor, the error model is derived from 12-h CGM recordings of two sensors used simultaneously and BG samples collected in parallel every 15 ± 5 min. Thanks to technological innovations, G4P outperforms 7P, with average mean absolute relative difference (MARD) of 11.1 versus 14.2%, respectively, and lowering of about 30% the error of each component. Thanks to the more sophisticated data processing algorithms, G4AP resulted more reliable than G4P, with a MARD of 10.0%, and a further decrease to 20% of the error due to blood-to-interstitial glucose kinetics.

Glucose variability indices in type 1 diabetes: parsimonious set of indices revealed by sparse principal component analysis

BACKGROUND

Continuous glucose monitoring (CGM) time-series are often analyzed, retrospectively, to investigate glucose variability (GV), a risk factor for the development of complications in type 1 diabetes (T1D). In the literature, several tens of different indices for GV quantification have been proposed, but many of them carry very similar information. The aim of this article is to select a relatively small subset of GV indices from a wider pool of metrics, to obtain a parsimonious but still comprehensive description of GV in T1D datasets.

MATERIALS AND METHODS

A pool of 25 GV indices was evaluated on two CGM time-series datasets of 17 and 16 T1D subjects, respectively, collected during the European Union Seventh Framework Programme project "Diadvisor" (2008-2012) in two different clinical research centers using the Dexcom(®) (San Diego, CA) SEVEN(®) Plus. After the indices were centered and scaled, the Sparse Principal Component Analysis (SPCA) technique was used to determine a reduced set of metrics that allows preserving a high percentage of the variance of the whole original set. In order to assess whether or not the selected subset of GV indices is dataset-dependent, the analysis was applied to both datasets, as well as to the one obtained by merging them.

RESULTS

SPCA revealed that a subset of up to 10 different GV indices can be sufficient to preserve more than the 60% of the variance originally explained by all the 25 variables. It is remarkable that four of these GV indices (i.e., Index of Glycemic Control, percentage of Glycemic Risk Assessment Diabetes Equation score due to euglycemia, percentage Coefficient of Variation, and Low Blood Glucose Index) were selected for all the considered T1D datasets.

CONCLUSIONS

The SPCA methodology appears a suitable candidate to identify, among the large number of literature GV indices, subsets that allow obtaining a parsimonious, but still comprehensive, description of GV.

Jump neural network for online short-time prediction of blood glucose from continuous monitoring sensors and meal information

Several real-time short-term prediction methods, based on time-series modeling of past continuous glucose monitoring (CGM) sensor data have been proposed with the aim of allowing the patient, on the basis of predicted glucose concentration, to anticipate therapeutic decisions and improve therapy of type 1 diabetes. In this field, neural network (NN) approaches could improve prediction performance handling in their inputs additional information. In this contribution we propose a jump NN prediction algorithm (horizon 30 min) that exploits not only past CGM data but also ingested carbohydrates information. The NN is tuned on data of 10 type 1 diabetics and then assessed on 10 different subjects. Results show that predictions of glucose concentration are accurate and comparable to those obtained by a recently proposed NN approach (Zecchin et al. (2012) [26]) having higher structural and algorithmical complexity and requiring the patient to announce the meals. This strengthen the potential practical usefulness of the new jump NN approach.

Physical activity measured by physical activity monitoring system correlates with glucose trends reconstructed from continuous glucose monitoring

BACKGROUND

In type 1 diabetes mellitus (T1DM), physical activity (PA) lowers the risk of cardiovascular complications but hinders the achievement of optimal glycemic control, transiently boosting insulin action and increasing hypoglycemia risk. Quantitative investigation of relationships between PA-related signals and glucose dynamics, tracked using, for example, continuous glucose monitoring (CGM) sensors, have been barely explored.

SUBJECTS AND METHODS

In the clinic, 20 control and 19 T1DM subjects were studied for 4 consecutive days. They underwent low-intensity PA sessions daily. PA was tracked by the PA monitoring system (PAMS), a system comprising accelerometers and inclinometers. Variations on glucose dynamics were tracked estimating first- and second-order time derivatives of glucose concentration from CGM via Bayesian smoothing. Short-time effects of PA on glucose dynamics were quantified through the partial correlation function in the interval (0, 60 min) after starting PA.

RESULTS

Correlation of PA with glucose time derivatives is evident. In T1DM, the negative correlation with the first-order glucose time derivative is maximal (absolute value) after 15 min of PA, whereas the positive correlation is maximal after 40-45 min. The negative correlation between the second-order time derivative and PA is maximal after 5 min, whereas the positive correlation is maximal after 35-40 min. Control subjects provided similar results but with positive and negative correlation peaks anticipated of 5 min.

CONCLUSIONS

Quantitative information on correlation between mild PA and short-term glucose dynamics was obtained. This represents a preliminary important step toward incorporation of PA information in more realistic physiological models of the glucose-insulin system usable in T1DM simulators, in development of closed-loop artificial pancreas control algorithms, and in CGM-based prediction algorithms for generation of hypoglycemic alerts.

Modeling the glucose sensor error

Continuous glucose monitoring (CGM) sensors are portable devices, employed in the treatment of diabetes, able to measure glucose concentration in the interstitium almost continuously for several days. However, CGM sensors are not as accurate as standard blood glucose (BG) meters. Studies comparing CGM versus BG demonstrated that CGM is affected by distortion due to diffusion processes and by time-varying systematic under/overestimations due to calibrations and sensor drifts. In addition, measurement noise is also present in CGM data. A reliable model of the different components of CGM inaccuracy with respect to BG (briefly, "sensor error") is important in several applications, e.g., design of optimal digital filters for denoising of CGM data, real-time glucose prediction, insulin dosing, and artificial pancreas control algorithms. The aim of this paper is to propose an approach to describe CGM sensor error by exploiting n multiple simultaneous CGM recordings. The model of sensor error description includes a model of blood-to-interstitial glucose diffusion process, a linear time-varying model to account for calibration and sensor drift-in-time, and an autoregressive model to describe the additive measurement noise. Model orders and parameters are identified from the n simultaneous CGM sensor recordings and BG references. While the model is applicable to any CGM sensor, here, it is used on a database of 36 datasets of type 1 diabetic adults in which n = 4 Dexcom SEVEN Plus CGM time series and frequent BG references were available simultaneously. Results demonstrates that multiple simultaneous sensor data and proper modeling allow dissecting the sensor error into its different components, distinguishing those related to physiology from those related to technology.

Signal processing algorithms implementing the "smart sensor" concept to improve continuous glucose monitoring in diabetes

Glucose readings provided by current continuous glucose monitoring (CGM) devices still suffer from accuracy and precision issues. In April 2013, we proposed a new conceptual architecture to deal with these problems and render CGM sensors algorithmically smarter, which consists of three modules for denoising, enhancement, and prediction placed in cascade to a commercial CGM sensor. The architecture was assessed on a data set consisting of 24 type 1 diabetes patients collected in four clinical centers of the AP@home Consortium (a European project of 7th Framework Programme funded by the European Committee). This article, as a companion to our prior publication, illustrates the technical details of the algorithms and of the implementation issues.

Non-invasive continuous glucose monitoring with multi-sensor systems: a Monte Carlo-based methodology for assessing calibration robustness

In diabetes research, non-invasive continuous glucose monitoring (NI-CGM) devices represent a new and appealing frontier. In the last years, some multi-sensor devices for NI-CGM have been proposed, which exploit several sensors measuring phenomena of different nature, not only for measuring glucose related signals, but also signals reflecting some possible perturbing processes (temperature, blood perfusion). Estimation of glucose levels is then obtained combining these signals through a mathematical model which requires an initial calibration step exploiting one reference blood glucose (RBG) sample. Even if promising results have been obtained, especially in hospitalized volunteers, at present the temporal accuracy of NI-CGM sensors may suffer because of environmental and physiological interferences. The aim of this work is to develop a general methodology, based on Monte Carlo (MC) simulation, to assess the robustness of the calibration step used by NI-CGM devices against these disturbances. The proposed methodology is illustrated considering two examples: the first concerns the possible detrimental influence of sweat events, while the second deals with calibration scheduling. For implementing both examples, 45 datasets collected by the Solianis Multisensor system are considered. In the first example, the MC methodology suggests that no further calibration adjustments are needed after the occurrence of sweat events, because the “Multisensor+model” system is able to deal with the disturbance. The second case study shows how to identify the best time interval to update the model's calibration for improving the accuracy of the estimated glucose. The methodology proposed in this work is of general applicability and can be helpful in making those incremental steps in NI-CGM devices development needed to further improve their performance.

Future Perspectives in Glucose Monitoring Sensors

The prevalence of diabetes is increasing. improved glucose control is fundamental to reduce both long-term micro- and macrovascular complications and short-term complications, such as diabetic ketoacidosis and severe hypoglycemia. Frequent blood glucose monitoring is an essential part of diabetes management. However, almost all available blood glucose monitoring devices are invasive. This determines a reduced patient compliance, which in turn reflects negatively on glucose control. Therefore, there is a need to develop noninvasive glucose monitoring devices that will reduce the need of invasive procedures, thus increasing patient compliance and consequently improving quality of life and health of patients with diabetes.