Search

We have attempted to estimate blood glucose levels based on facial images measured in the near-infrared band, which is highly biopermeable, to establish a remote minimally invasive blood glucose measurement method. We measured facial images in the near-infrared wavelength range of 760–1650 nm, and constructed a general model for blood glucose level estimation by linear regression using the weights of spatial features of the measured facial images as explanatory variables. The results showed that the accuracy values of blood glucose estimation in the generalization performance evaluation were 43.02 mg/dL for NIR-I (760–1100 nm) and 43.61 mg/dL for NIR-II (1050–1650 nm) in the RMSE of the general model. Since biological information is nonlinear, it is necessary to explore suitable modeling methods for blood glucose estimation, including not only linear regression but also nonlinear regression. The purpose of this study is to explore suitable regression methods among linear and nonlinear regression methods to construct a blood glucose estimation model based on facial images with wavelengths from 760 to 1650 nm. The results showed that model using Random Forest had the best estimation accuracy with an RMSE of 36.02 mg/dL in NIR-I and the MR model had the best estimation accuracy with RMSE of 36.70 mg/dL in NIR-II under the current number of subjects and measurement data points. The independent components selected for the model have spatial features considered to be simply individual differences that are not related to blood glucose variation. [ABSTRACT FROM AUTHOR]

Simple Summary: Although some of the novel systemic treatments, especially the group of tyrosine kinase inhibitors, have shown a durable central nervous system response, ionizing radiation remains the mainstay in the management of brain metastases (BM). Recent technological advancements have enabled the replacement of whole-brain radiotherapy with localized stereotactic radiotherapy (SRT) for treating up to 10 BM, either as a primary or combined treatment, reducing neurotoxicity and improving local control (LC). The delivered target dose and patient selection play a crucial role in enhancing treatment efficacy. However, there is still limited evidence supporting which factors most affect LC and which patients derive the greatest benefit from SRT. This retrospective single-institutional study evaluated treatment outcomes in a heterogeneous patient population treated with Linac-based SRT, with the aim of identifying potential dosimetric and clinical prognostic factors to better inform the decision-making process. Background/Objectives: To report on predictive factors in Linac-based SRT for single and multiple BM. Methods: Consecutive patients receiving either one or three fractions of single-isocenter coplanar VMAT SRT were retrospectively included. The GTV-PTV margin was 1–2 mm. The delivered target dose was estimated by recalculating the original plans on roto-translated CT according to errors recorded by post-treatment CBCT. The Kaplan–Meier method estimated local progression-free survival (LPFS), intracranial progression-free survival (IPFS), and overall survival (OS). Log-rank and Wilcoxon–Mann–Whitney tests evaluated inter-group differences, whereas Cox regression analysis assessed prognostic factors. Results: Fifty females and fifty males, with a median age of 69 years, received 107 SRTs. A total of 213 BM (range, 1–10 per treatment) with a median volume of 0.22 cc were irradiated with a median minimum BED of 59.5 Gy. The median delivered GTV D95 reduction was −0.3%. The median follow-up was 11 months. Nineteen LP events and a 1-year LC rate of 90.1% were observed. The GTV coverage did not correlate with LC, while the GTV volume was a risk factor for LP, with the 1-year rate dropping to 73% for volumes ≥ 0.88 cc. The median LPFS, IPFS, and OS were 6, 5, and 7 months, respectively. Multivariate analysis showed that patients with melanoma histology and those receiving a second or subsequent systemic therapy line had the worst outcomes, whereas patients with adenocarcinoma histology and mutations showed better results. Conclusions: The accuracy and efficacy of the Linac-based SRT approach for BM were confirmed, but the dose distribution alone failed to predict the treatment response, suggesting that other factors must be considered to maximize SRT outcomes. [ABSTRACT FROM AUTHOR]

Simple Summary: Distinguishing radionecrosis from local recurrence after cranial radiotherapy remains a challenging diagnostic dilemma due to their overlapping clinical manifestations. Although new MRI techniques and metabolic imaging have been developed, diagnostic performance remains variable. Recently, our institution acquired a PET-MRI, enabling us to investigate the diagnostic performance of multiparametric MRI when used in dual-phase 18F-FDG PET-mpMRI to enhance diagnostic accuracy. Brain metastases or primary brain tumours had poor prognosis until the use of high dose radiotherapy. However, radionecrosis is a complex challenge in the post-radiotherapy management of these patients due to the difficulty of distinguishing this complication from local tumour recurrence. MRI alone has a variable specificity and sensibility, as does PET-CT imaging. We aimed to investigate the diagnostic performance of dual-phase 18F-FDG PET-mpMRI to distinguish cerebral radionecrosis from local tumour recurrence after cranial radiotherapy. A retrospective analysis was conducted between May 2021 and September 2022. Inclusion criteria encompassed patients with inconclusive MRI findings post-radiotherapy and history of cerebral radiotherapy for primary or metastatic brain lesions. Lesions are assessed qualitatively and semi-quantitatively. The gold standard to assess radionecrosis was histopathology or a composite criterion at three months. The study evaluated 24 lesions in 23 patients. Qualitative analysis yielded 85.7% sensitivity and 75% specificity. Semi-quantitative analysis, based on contralateral background noise, achieved 100% sensitivity and 50% specificity. Moreover, using contralateral frontal lobe background noise resulted in higher performances with 92% sensitivity and 63% specificity. Stratification by lesion type demonstrated 100% sensitivity and specificity rates for metastatic lesions. The diagnostic performance of dual-phase 18F-FDG PET-mpMRI shows promising results for metastatic lesions. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

In hybrid automatic insulin delivery (HAID) systems, meal disturbance is compensated by feedforward control, which requires the announcement of the meal by the patient with type 1 diabetes (DM1) to achieve the desired glycemic control performance. The calculation of insulin bolus in the HAID system is based on the amount of carbohydrates (CHO) in the meal and patient-specific parameters, i.e. carbohydrate-to-insulin ratio (CR) and insulin sensitivity-related correction factor (CF). The estimation of CHO in a meal is prone to errors and is burdensome for patients. This study proposes a fully automatic insulin delivery (FAID) system that eliminates patient intervention by compensating for unannounced meals. This study exploits the deep reinforcement learning (DRL) algorithm to calculate insulin bolus for unannounced meals without utilizing the information on CHO content. The DRL bolus calculator is integrated with a closed-loop controller and a meal detector (both previously developed by our group) to implement the FAID system. An adult cohort of 68 virtual patients based on the modified UVa/Padova simulator was used for in-silico trials. The percentage of the overall duration spent in the target range of 70–180 mg/dL was 71.2 % and 76.2 % , < 70 mg/dL was 0.9 % and 0.1 % , and > 180 mg/dL was 26.7 % and 21.1 % , respectively, for the FAID system and HAID system utilizing a standard bolus calculator (SBC) including CHO misestimation. The proposed algorithm can be exploited to realize FAID systems in the future. [ABSTRACT FROM AUTHOR]

Along with the developing technology, obesity and diabetes are increasing rapidly among people. The identification of diabetes diseases, modeling, predicting their behavior, conducting simulations, studying control and treatment methods using mathematical methods has become of great importance. In this paper, we have obtained numerical solutions by considering the glucose-insulin fractional model. This model consists of three compartments: the blood glucose concentration (G), the blood insulin concentration (I) and the ready-to-absorb glucose concentration (D) in the small intestine. The fractional derivative is used in the sense of Caputo. By performing mathematical analyzes for the Glucose-Insulin fractional mathematical model, numerical results were obtained with the help of the Euler method and graphs were drawn. [ABSTRACT FROM AUTHOR]

Regulating blood glucose level (BGL) for type-1 diabetic patient (T1DP) accurately is very important issue, an uncontrolled BGL outside the standard safe range between 70 and 180 mg/dl results in dire consequences for health and can significantly increase the chance of death. So the purpose of this study is to design an optimized controller that infuses appropriate amounts of exogenous insulin into the blood stream of T1DP proportional to the amount of obtained glucose from food. The nonlinear extended Bergman minimal model is used to present glucose-insulin physiological system, an interval type-2 fuzzy logic controller (IT2FLC) is utilized to infuse the proper amount of exogenous insulin. Superiority of IT2FLC in minimizing the effect of uncertainties in the system depends primarily on the best choice of footprint of uncertainty (FOU) of IT2FLC. So a comparison includes four different optimization methods for tuning FOU including hybrid grey wolf optimizer-cuckoo search (GWOCS) and fuzzy logic controller (FLC) method is constructed to select the best controller approach. The effectiveness of the proposed controller was evaluated under six different scenarios of T1DP using Matlab/Simulink platform. A 24-h scenario close to real for 100 virtual T1DPs subjected to parametric uncertainty, uncertain meal disturbance and random initial condition showed that IT2FLC accurately regulate BGL for all T1DPs within the standard safe range. The results indicated that IT2FLC using GWOCS can prevent side effect of treatment with blood-sugar-lowering medication. Also stability analysis for the system indicated that the system operates within the stability region of nonlinear system. [ABSTRACT FROM AUTHOR]

Macronutrients play a crucial role in management of type 2 diabetes. This is seen in their ability to modulate plasma glucose concentrations. However, the ideal proportions of macronutrients to be consumed in order to maintain ideal plasma glucose concentrations remains elusive. Therefore, this paper set out to conduct a review to investigate the relationship between macronutrients and plasma glucose concentrations from a physiological perspective. The review was conducted using papers obtained from various databases such as MEDLINE (Pubmed), Open Access Journals Elsevier, Free Medical Journals and Google Scholar. The research papers included general reviews, systemic reviews, meta-analyses, and randomized control trials that examined the effect of macronutrients on plasma glucose concentration as well as papers on mathematical models describing the relationship between macronutrient bioavailability and plasma glucose concentration. The review assessed the effect of various macronutrients on postprandial plasma glucose concentration, post-prandial plasma insulin, post-prandial glucose-dependent insulinotropic peptide plasma concentration, and post-prandial glucose-like peptide-1 plasma concentration. The results of the review showed that carbohydrates influence plasma glucose concentration in a dose dependant manner but this is dependent on their bioavailability. This bioavailability was shown to be subject to fluctuations determined by food processing techniques, food structure, and food matrix. The results also showed that some specific types of fats and proteins indirectly influence plasma glucose concentration through their effect on incretin hormones. The effect of fats and proteins on incretin hormones was through different mechanisms and pathways. In-lieu of the findings, the review concludes that the macronutrient composition of diets designed for type 2 diabetic patients should take into consideration the physiological relationship between the macronutrients and plasma glucose concentrations. In this way, diet proportions can be made in such a manner as to determine the exact amounts that will realize near normal plasma glucose concentrations for a type 2 diabetic patient. [ABSTRACT FROM AUTHOR]

People with type 1 diabetes are required to adhere to their treatment rigorously to ensure maximum benefits. Diabetes tracking tools have played an important role in this regard. Type 1 diabetes monitoring has evolved and matured with the advent of blood glucose monitor sensors, insulin pens, and insulin pump automation. However, carbohydrate monitoring has seen little progress despite carbohydrates representing a major potential disruption. Relying on the modeling of carbohydrate intake using the rate of exogenous glucose appearance, we first present a methodology capable of identifying the type of carbohydrates ingested by classifying them into fast and non-fast carbohydrates. Second, we test the ability of the methodology to identify the correct synchrony between the actual mealtime and the time labeled as such in diabetes records. A deep neural network is trained with processed input data that consist of different values to estimate the parameters in a series of experiments in which, firstly, we vary the response of ingested carbohydrates for subsequent identification and, secondly, we shift the learned carbohydrate absorption curves in time to estimate when the meals were administered to virtual patients. This study validates that the identification of different carbohydrate classes in the meal records of people with type 1 diabetes could become a valuable source of information, as it demonstrates the potential to identify inaccuracies in the recorded meal records of these patients, suggesting the potential abilities of the next generation of type 1 diabetes management tools. [ABSTRACT FROM AUTHOR]

Patients with type 1 diabetes must continually decide how much insulin to inject before each meal to maintain blood glucose levels within a healthy range. Recent research has worked on a solution for this burden, showing the potential of reinforcement learning as an emerging approach for the task of controlling blood glucose levels. In this paper, we test and evaluate several deep Q-learning algorithms for automated and personalized blood glucose regulation in an in silico type 1 diabetes patient with the goal of estimating and delivering proper insulin doses. The proposed algorithms are model-free approaches with no prior information about the patient. We used the Hovorka model with meal variation and carbohydrate counting errors to simulate the patient included in this work. Our experiments compare different deep Q-learning extensions showing promising results controlling blood glucose levels, with some of the proposed algorithms outperforming standard baseline treatment. [ABSTRACT FROM AUTHOR]

Type I diabetes mellitus is a serious autoimmune condition impacting a large population around the world that need a daily infusion of insulin substitutes to regulate blood glucose levels within healthy limits. The purpose of the study was to design a robust μ -controller based on an uncertain linear-time invariant (LTI) representation of the Hovorka model for glucose–insulin metabolism. The model set was obtained using linearization around an equilibrium point and adding parametric uncertainty to account for the time delay variation between plasma glucose concentration and its subcutaneous measurement. As a result, the robust stability and performance of the closed loop were proved using the structured singular value μ. The performance of the designed controller was also checked with a numerical simulation in connection with the nonlinear model. [ABSTRACT FROM AUTHOR]

Comparing the technical reliability and insulin dosing of a "do-it-yourself artificial pancreas" with a commercial hybrid closed-loop system in a "shadow-mode" scenario: An exploratory study Day-and-night glycaemic control with closed-loop insulin delivery versus conventional insulin pump therapy in free-living adults with well controlled type 1 diabetes: an open-label, randomised, crossover study. Keywords: CSII; insulin pump therapy; subcutaneous injection; type 1 diabetes EN CSII insulin pump therapy subcutaneous injection type 1 diabetes 2780 2783 4 08/03/23 20230901 NES 230901 INTRODUCTION Do-it-yourself artificial pancreas systems (DIYAPSs) are compound devices that autonomously adjust insulin dosage in people with diabetes. However, the DIYAPS was not delivering insulin, which might have caused an overcompensation of insulin adaptions if delivered insulin doses were lower than calculated by the DIYAPS. [Extracted from the article]

In the last decade, artificial intelligence (AI) and AI-mediated technologies have undergone rapid evolution in healthcare and medicine, from apps to computer software able to analyze medical images, robotic surgery and advanced data storage system. The main aim of the present commentary is to briefly describe the evolution of AI and its applications in healthcare, particularly in nutrition and clinical biochemistry. Indeed, AI is revealing itself to be an important tool in clinical nutrition by using telematic means to self-monitor various health metrics, including blood glucose levels, body weight, heart rate, fat percentage, blood pressure, activity tracking and calorie intake trackers. In particular, the application of the most common digital technologies used in the field of nutrition as well as the employment of AI in the management of diabetes and obesity, two of the most common nutrition-related pathologies worldwide, will be presented. [ABSTRACT FROM AUTHOR]

In view of the poor timeliness of dynamic blood glucose data and a delay of insulin effect for blood glucose control, and considering the nonlinearity and nonstationarity of the glucose data, a new blood Glucose Prediction algorithm combined Correlation coefficient-based complete ensemble empirical mode decomposition with adaptive noise and back propagation neural network (GPCEMBP) was proposed to increase the prediction time and improve the prediction accuracy. It refined the mode decomposition algorithm and integrated the correlation mode filter function to extract the characteristic intrinsic mode functions from the original signal. A new neural network prediction model was constructed by optimizing the number of hidden layer neurons, the number of hidden layers, activation functions, the number of inputs, structure, and other parameters. Finally, a predicted blood glucose value was calculated by phase space reconstruction technology. Through ablation and comparison experiments, it was demonstrated that the GPCEMBP algorithm had better prediction accuracy, convergence, and robustness in blood glucose prediction within 84 min. In addition, it has good adaptability to deal with different quality glucose data. [ABSTRACT FROM AUTHOR]

Mathematical modeling of mass transfer and absorption in the small intestine has been a challenging task. Systematic review and analysis of existing efforts indicate the need to pursue a reliable predictive model that is physically sound and computationally efficient. With the consideration of 3D intestinal inner wall structure, this work rigorously derives an absorption model that can be used as a source term in a 1D distributed model, conventionally called the diffusion–convection–reaction model. Moreover, computational fluid dynamics simulations are carried out to generate in silico experimental data for quantification of the mass‐transfer coefficient in the absorption model. This model facilitates a better understanding of the intricate influence of intestinal morphology and motility on mass transfer and absorption in the intestine. Rat duodenum featuring a villous structure and pendular movement is selected as an example to demonstrate the capability of this approach. [ABSTRACT FROM AUTHOR]

Most advanced technologies for the treatment of type 1 diabetes, such as sensor-pump integrated systems or the artificial pancreas, require accurate glucose predictions on a given future time-horizon as a basis for decision-making support systems. Seasonal stochastic models are data-driven algebraic models that use recent history data and periodic trends to accurately estimate time series data, such as glucose concentration in diabetes. These models have been proven to be a good option to provide accurate blood glucose predictions under free-living conditions. These models can cope with patient variability under variable-length time-stamped daily events in supervision and control applications. However, the seasonal-models-based framework usually needs of several months of data per patient to be fed into the system to adequately train a personalized glucose predictor for each patient. In this work, an in silico analysis of the accuracy of prediction is presented, considering the effect of training a glucose predictor with data from a cohort of patients (population) instead of data from a single patient (individual). Feasibility of population data as an input to the model is asserted, and the effect of the dataset size in the determination of the minimum amount of data for a valid training of the models is studied. Results show that glucose predictors trained with population data can provide predictions of similar magnitude as those trained with individualized data. Overall median root mean squared error (RMSE) (including 25% and 75% percentiles) for the predictor trained with population data are { 6.96 [ 4.87 , 8.67 ] , 12.49 [ 7.96 , 14.23 ] , 19.52 [ 10.62 , 23.37 ] , 28.79 [ 12.96 , 34.57 ] , 32.3 [ 16.20 , 41.59 ] , 28.8 [ 15.13 , 37.18 ] } mg/dL, for prediction horizons (PH) of { 15 , 30 , 60 , 120 , 180 , 240 } min, respectively, while the baseline of the individually trained RMSE results are { 6.37 [ 5.07 , 6.70 ] , 11.27 [ 8.35 , 12.65 ] , 17.44 [ 11.08 , 20.93 ] , 22.72 [ 14.29 , 28.19 ] , 28.45 [ 14.79 , 34.38 ] , 25.58 [ 13.10 , 36.60 ] } mg/dL, both training with 16 weeks of data. Results also show that the use of the population approach reduces the required training data by half, without losing any prediction capability. [ABSTRACT FROM AUTHOR]

This paper proposes a simulation linear model created in the MATLAB environment, which provides a process for regulating blood sugar levels. The controller is built for the need for any type of diabetes to control and normalize the blood sugar content of the patient in order to eliminate the differences in the quality of life of a diabetic patient and a healthy person. The linearization of the nonlinear model was performed, and the adequacy of the linearized model was verified and confirmed using the MATLAB simulation. The choice of the PID controller and the CHR method for its adjustment was justified and MATLAB tools were used to show the implementation of these methods. The model of the patient with the controller has been built; the algorithm for the automatic adjustment of the PID controller parameters has been developed and realized. The directions for continuation of the work on this problem regarding regulation in the system under study are proposed. [ABSTRACT FROM AUTHOR]

In this paper, we propose an innovative Federated Learning-inspired evolutionary framework. Its main novelty is that this is the first time that an Evolutionary Algorithm is employed on its own to directly perform Federated Learning activity. A further novelty resides in the fact that, differently from the other Federated Learning frameworks in the literature, ours can efficiently deal at the same time with two relevant issues in Machine Learning, i.e., data privacy and interpretability of the solutions. Our framework consists of a master/slave approach in which each slave contains local data, protecting sensible private data, and exploits an evolutionary algorithm to generate prediction models. The master shares through the slaves the locally learned models that emerge on each slave. Sharing these local models results in global models. Being that data privacy and interpretability are very significant in the medical domain, the algorithm is tested to forecast future glucose values for diabetic patients by exploiting a Grammatical Evolution algorithm. The effectiveness of this knowledge-sharing process is assessed experimentally by comparing the proposed framework with another where no exchange of local models occurs. The results show that the performance of the proposed approach is better and demonstrate the validity of its sharing process for the emergence of local models for personal diabetes management, usable as efficient global models. When further subjects not involved in the learning process are considered, the models discovered by our framework show higher generalization capability than those achieved without knowledge sharing: the improvement provided by knowledge sharing is equal to about 3.03% for precision, 1.56% for recall, 3.17% for F 1 , and 1.56% for accuracy. Moreover, statistical analysis reveals the statistical superiority of model exchange with respect to the case of no exchange taking place. [ABSTRACT FROM AUTHOR]

Accurate treatment adjustment to physical activity (PA) remains a challenging problem in type 1 diabetes (T1D) management. Exercise-driven effects on glucose metabolism depend strongly on duration and intensity of the activity, and are highly variable between patients. In-silico evaluation can support the development of improved treatment strategies, and can facilitate personalized treatment optimization. This requires models of the glucose-insulin system that capture relevant exercise-related processes. We developed a model of glucose-insulin regulation that describes changes in glucose metabolism for aerobic moderate- to high-intensity PA of short and prolonged duration. In particular, we incorporated the insulin-independent increase in glucose uptake and production, including glycogen depletion, and the prolonged rise in insulin sensitivity. The model further includes meal absorption and insulin kinetics, allowing simulation of everyday scenarios. The model accurately predicts glucose dynamics for varying PA scenarios in a range of independent validation data sets, and full-day simulations with PA of different timing, duration and intensity agree with clinical observations. We personalized the model on data from a multi-day free-living study of children with T1D by adjusting a small number of model parameters to each child. To assess the use of the personalized models for individual treatment evaluation, we compared subject-specific treatment options for PA management in replay simulations of the recorded data with altered meal, insulin and PA inputs. Author summary: Exercise represents a cornerstone of diabetes management. Yet, many people with type 1 diabetes refrain from exercising, since it increases the risk for hypoglycemia and requires adjusted insulin treatment. The effect of exercise on blood glucose levels depends on exercise duration and intensity, but also varies strongly between individuals, making accurate adjustment a challenge. Mathematical models can help to better understand exercise physiology and to devise new treatment strategies. Here, we propose a model of glucose-insulin regulation that captures the effects of exercise on glucose metabolism and personalize it to individual children with type 1 diabetes, allowing subject-specific treatment assessment. [ABSTRACT FROM AUTHOR]

Improving the prediction of blood glucose concentration may improve the quality of life of people living with type 1 diabetes by enabling them to better manage their care. Given the anticipated benefits of such a prediction, numerous methods have been proposed. Rather than attempting to predict glucose concentration, a deep learning framework for prediction is proposed in which prediction is performed using a scale for hypo- and hyper-glycemia risk. Using the blood glucose risk score formula proposed by Kovatchev et al., models with different architectures were trained, including, a recurrent neural network (RNN), a gated recurrent unit (GRU), a long short-term memory (LSTM) network, and an encoder-like convolutional neural network (CNN). The models were trained using the OpenAPS Data Commons data set, comprising 139 individuals, each with tens of thousands of continuous glucose monitor (CGM) data points. The training set was composed of 7% of the data set, while the remaining was used for testing. Performance comparisons between the different architectures are presented and discussed. To evaluate these predictions, performance results are compared with the last measurement (LM) prediction, through a sample-and-hold approach continuing the last known measurement forward. The results obtained are competitive when compared to other deep learning methods. A root mean squared error (RMSE) of 16 mg/dL, 24 mg/dL, and 37 mg/dL were obtained for CNN prediction horizons of 15, 30, and 60 min, respectively. However, no significant improvements were found for the deep learning models compared to LM prediction. Performance was found to be highly dependent on architecture and the prediction horizon. Lastly, a metric to assess model performance by weighing each prediction point error with the corresponding blood glucose risk score is proposed. Two main conclusions are drawn. Firstly, going forward, there is a need to benchmark model performance using LM prediction to enable the comparison between results obtained from different data sets. Secondly, model-agnostic data-driven deep learning models may only be meaningful when combined with mechanistic physiological models; here, it is argued that neural ordinary differential equations may combine the best of both approaches. These findings are based on the OpenAPS Data Commons data set and are to be validated in other independent data sets. [ABSTRACT FROM AUTHOR]

In silico simulation studies using 5807 virtual patients with insulin dependent diabetes have been conducted to estimate the risk and efficacy with the closed-loop 780G pump when switching between Medtronic Guardian Sensor 3 (GS3) and Medtronic Guardian 4 Sensor (G4S), next generation calibration free glucose sensor. To demonstrate by utilizing a case study that captures the merits of in silico studies with single hormone insulin dependent virtual patients that include variability in pharmacokinetics/pharmacodynamics, age, gender, insulin sensitivity and BMIs. Also, to show that in silico studies can uniquely isolate the effect of a single variable on clinical outcomes. Simulation studies results were compared to clinical and commercial data and were separated by age groups and pump settings. The commercial data, the clinical study data and the simulation studies predicted that switching between GS3 to G4S will introduce a change in glucose average, percentage time between 70 and 180 mg/dL, and percentage time below 70 mg/dL of: 5.2, 3.4, and 3.1 mg/dL, − 1.1, 0.2, and − 1.1%, and − 0.6, − 1.0, and − 0.3%, respectively. We demonstrated that our simulation studies were able to predict the difference in glycemic outcomes when switching between different sensors in real world setting, better than a small clinical controlled study. As predicted, switching between GS3 and G4S sensors with the 780G system does not introduce clinical risk and maintain the clinical outcomes of the sensor. We demonstrated the ability of insulin dependent diabetes virtual patients to predict clinical outcomes and to augment or even replace some small clinical studies. [ABSTRACT FROM AUTHOR]

The article presents the discussion on inaugural guideline on diabetes technology being published. Topics include created a means to utilize technology without requiring on body devices though studies in the pediatric population; and clinical trials and real‐world data demonstrated improvements in glycemia with use of automated insulin delivery (AID).

Type 1 diabetes mellitus is a disease that affects millions of people around the world. Recent progress in embedded devices has allowed the development of artificial pancreas that can pump insulin subcutaneously to automatically regulate blood glucose levels in diabetic patients. In this work, a Lyapunov-based intelligent controller using artificial neural networks is proposed for application in automated insulin delivery systems. The adoption of an adaptive radial basis function network within the control scheme allows regulation of blood glucose levels without the need for a dynamic model of the system. The proposed model-free approach does not require the patient to inform when they are going to have a meal and is able to deal with inter- and intrapatient variability. To ensure safe operating conditions, the stability of the control law is rigorously addressed through a Lyapunov-like analysis. In silico analysis using virtual patients are provided to demonstrate the effectiveness of the proposed control scheme, showing its ability to maintain normoglycemia in patients with type 1 diabetes mellitus. Three different scenarios were considered: one long- and two short-term simulation studies. In the short-term analyses, 20 virtual patients were simulated for a period of 7 days, with and without prior basal therapy, while in the long-term simulation, 1 virtual patient was assessed over 63 days. The results show that the proposed approach was able to guarantee a time in the range above 95% for the target glycemia in all scenarios studied, which is in fact well above the desirable 70%. Even in the long-term analysis, the intelligent control scheme was able to keep blood glucose metrics within clinical care standards: mean blood glucose of 119.59 mg/dL with standard deviation of 32.02 mg/dL and coefficient of variation of 26.78%, all below the respective reference values. [ABSTRACT FROM AUTHOR]

Automated insulin delivery (AID) systems, known as artificial pancreas or closed-loop glucose control systems, have been developed to improve the glycemic outcomes of people with type 1 diabetes. These systems use a control algorithm that automatically modifies the amount of insulin infused into a patient based on real-time blood glucose measurements. This study presents a summary of key clinical and technical issues related to the development of the first commercial AID systems and their evolution into commercial biomedical devices. Highlights of each AID system are summarized through timelines, ranging from the definition of the core strategy of the control algorithm to the practical application and subsequent commercial approval. Tabulated information regarding the conducted main clinical studies is also presented. Insulin therapy has evolved up to the current commercial AID systems available, which have provided patients access to a safer and more effective therapy owing to automatic adjustments to insulin. However, this technology is relatively new and can be significantly improved. Limitations include the resistance of healthcare providers, high costs, and the availability of this treatment. The future of this technology is directed toward obtaining fully automatic control systems. [ABSTRACT FROM AUTHOR]

Mathematical modeling of the glucose–insulin system forms the core of simulators in the field of glucose metabolism. The complexity of human biological systems makes it a challenging task for the physiological models to encompass the entirety of such systems. Even though modern diabetes simulators perform a respectable task of simulating the glucose–insulin action, they are unable to estimate various phenomena affecting the glycemic profile of an individual such as glycemic disturbances and patient behavior. This research work presents a potential solution to this problem by proposing a method for the generation of blood glucose values conditioned on plasma insulin approximation of type 1 diabetes patients using a pixel-to-pixel generative adversarial network. Two type-1 diabetes cohorts comprising 29 and 6 patients, respectively, are used to train the generative model. This study shows that the generated blood glucose values are statistically similar to the real blood glucose values, mimicking the time-in-range results for each of the standard blood glucose ranges in type 1 diabetes management and obtaining similar means and variability outcomes. Furthermore, the causal relationship between the plasma insulin values and the generated blood glucose conforms to the same relationship observed in real patients. These results herald the aptness of deep generative models for the generation of virtual patients with diabetes. [ABSTRACT FROM AUTHOR]

A key ingredient in approximate Bayesian computation (ABC) procedures is the choice of a discrepancy that describes how different the simulated and observed data are, often based on a set of summary statistics when the data cannot be compared directly. Unless discrepancies and summaries are available from experts or prior knowledge, which seldom occurs, they have to be chosen, and thus their choice can affect the quality of approximations. The choice between discrepancies is an active research topic, which has mainly considered data discrepancies requiring samples of observations or distances between summary statistics. In this work, we introduce a preliminary learning step in which surrogate posteriors are built from finite Gaussian mixtures using an inverse regression approach. These surrogate posteriors are then used in place of summary statistics and compared using metrics between distributions in place of data discrepancies. Two such metrics are investigated: a standard L 2 distance and an optimal transport-based distance. The whole procedure can be seen as an extension of the semi-automatic ABC framework to a functional summary statistics setting and can also be used as an alternative to sample-based approaches. The resulting ABC quasi-posterior distribution is shown to converge to the true one, under standard conditions. Performance is illustrated on both synthetic and real data sets, where it is shown that our approach is particularly useful when the posterior is multimodal. [ABSTRACT FROM AUTHOR]

Objective: The interpretation of time series data collected in free-living has gained importance in chronic disease management. Some data are collected objectively from sensors and some are estimated and entered by the individual. In type 1 diabetes (T1D), blood glucose concentration (BGC) data measured by continuous glucose monitoring (CGM) systems and insulin doses administered can be used to detect the occurrences of meals and physical activities and generate the personal daily living patterns for use in automated insulin delivery (AID). Methods: Two challenges in time-series data collected in daily living are addressed: data quality improvement and the detection of unannounced disturbances of BGC. CGM data have missing values for varying periods of time and outliers. People may neglect reporting their meal and physical activity information. In this work, novel methods for preprocessing real-world data collected from people with T1D and the detection of meal and exercise events are presented. Four recurrent neural network (RNN) models are investigated to detect the occurrences of meals and physical activities disjointly or concurrently. Results: RNNs with long short-term memory (LSTM) with 1D convolution layers and bidirectional LSTM with 1D convolution layers have average accuracy scores of 92.32% and 92.29%, and outperform other RNN models. The F1 scores for each individual range from 96.06% to 91.41% for these two RNNs. Conclusions: RNNs with LSTM and 1D convolution layers and bidirectional LSTM with 1D convolution layers provide accurate personalized information about the daily routines of individuals. Significance: Capturing daily behavior patterns enables more accurate future BGC predictions in AID systems and improves BGC regulation. [ABSTRACT FROM AUTHOR]

The computational burden of iterative online optimization–based model predictive control (MPC) process is solved by adapting off-line optimization-based nonlinear explicit model predictive control (NEMPC). In this paper, the application of NEMPC is verified to regulate blood glucose level in type 1 diabetes mellitus (T1DM) patients. The objective of glucose regulation is to avoid hyperglycemia (> 180 mg/dl) and hypoglycemia (< 50 mg/dl) by maintaining glucose level in the range of 70 to 180 mg/dl. It helps to avoid time complexity of iterative process during solution of optimization stage in MPC. In the nonlinear T1DM model, only the state dynamic of sugar is measurable with low complexity and high cost among other states of the model. Therefore, an extended Kalman filter (EKF) is used developed to estimate unavailable states. The information of the estimated states are used to develop the proposed control approach for the T1DM patients. The simulation results of the NEMPC along with EKF-based state estimator for T1DM model shows the regulation of blood glucose-level (BGL) within 70 to 180 mg/dl within 120 min. The robustness of the proposed scheme is also verified under the change in parameters and food disturbance. The control variability grid analysis (CVGA) of NEMPC for 50 numbers of virtual T1DM patients under random parametric changes and meal disturbances shows the avoidance of hypoglycemia and hyperglycemia in type 1 diabetic patients. The proposed control method is simple, robust and efficient to regulate glucose level in T1DM patients. [ABSTRACT FROM AUTHOR]

We investigate the use of learning approaches to handle Bayesian inverse problems in a computationally efficient way when the signals to be inverted present a moderately high number of dimensions and are in large number. We propose a tractable inverse regression approach which has the advantage to produce full probability distributions as approximations of the target posterior distributions. In addition to provide confidence indices on the predictions, these distributions allow a better exploration of inverse problems when multiple equivalent solutions exist. We then show how these distributions can be used for further refined predictions using importance sampling, while also providing a way to carry out uncertainty level estimation if necessary. The relevance of the proposed approach is illustrated both on simulated and real data in the context of a physical model inversion in planetary remote sensing. The approach shows interesting capabilities both in terms of computational efficiency and multimodal inference. [ABSTRACT FROM AUTHOR]

Recent reviews conclude that aphasia intervention is effective. However, replication and implementation require detailed reporting of intervention is and a specification of participant profiles. To date, reviews concentrate more on efficacy than on intervention reporting quality. The aim of this project is to review the descriptions of aphasia interventions and participants appearing in recent systematic reviews of aphasia intervention effectiveness. The relationship between the quality of these descriptions and the robustness of research design is explored, and the replicability of aphasia interventions is evaluated. The scope of our search was an analysis of the aphasia intervention studies included in the and EBRSR 2018 systematic reviews, and in the RCSLT 2014 literature synthesis. Intervention descriptions published separately from the intervention study (i.e. published online, in clinical tools, or a separate trial protocols) were not included. The criteria for inclusion were that participants had aphasia, the intervention involved language and/or communication, and included the following research designs: Randomised Controlled Trial (RCT), comparison or control, crossover design, case series. Exclusion criteria included non-SLT interventions, studies involving fewer than four participants, conference abstracts, studies not available in English. Studies were evaluated for completeness of intervention description using the TIDieR Checklist. Additionally, we rated the quality of patient and intervention description, with particular reference to replicability. Ninety-three studies were included. Only 14 studies (15%) had >50 participants. Fifty-six studies (60%) did not select participants with a specific aphasia profile, and a further 10 studies only described participants as non-fluent. Across the studies, an average of eight (of 12) TIDieR checklist items were given but information on where, tailoring, modification and fidelity items was rarely available. Studies that evaluated general aphasia intervention approaches tended to use RCT designs, whereas more specific intervention studies were more likely to use case series designs. : Group studies were generally under-powered and there was a paucity of research looking at specific aphasia interventions for specific aphasia profiles. There was a trade-off between the robustness of the design and the level of specificity of the intervention described. While the TIDieR framework is a useful guide to information which should be included in an intervention study, it is insufficiently sensitive for assessing replicability. We consider possible solutions to the challenges of making large-scale trials more useful for determining effective aphasia intervention. [ABSTRACT FROM AUTHOR]

This paper proposes a robust control design method using reinforcement learning for controlling partially-unknown dynamical systems under uncertain conditions. The method extends the optimal reinforcement learning algorithm with a new learning technique based on the robust control theory. By learning from the data, the algorithm proposes actions that guarantee the stability of the closed-loop system within the uncertainties estimated also from the data. Control policies are calculated by solving a set of linear matrix inequalities. The controller was evaluated using simulations on a blood glucose model for patients with Type 1 diabetes. Simulation results show that the proposed methodology is capable of safely regulating the blood glucose within a healthy level under the influence of measurement and process noises. The controller has also significantly reduced the post-meal fluctuation of the blood glucose. A comparison between the proposed algorithm and the existing optimal reinforcement learning algorithm shows the improved robustness of the closed-loop system using our method. [ABSTRACT FROM AUTHOR]

Due to the sensitive nature of diabetes-related data, preventing them from being easily shared between studies, and the wide discrepancies in their data processing pipeline, progress in the field of glucose prediction is hard to assess. To address this issue, we introduce GLYFE (GLYcemia Forecasting Evaluation), a benchmark of machine learning-based glucose predictive models. We present the accuracy and clinical acceptability of nine different models coming from the literature, from standard autoregressive to more complex neural network-based models. These results are obtained on two different datasets, namely UVA/Padova Type 1 Diabetes Metabolic Simulator (T1DMS) and Ohio Type-1 Diabetes Mellitus (OhioT1DM), featuring artificial and real type 1 diabetic patients respectively. By providing extensive details about the data flow as well as by providing the whole source code of the benchmarking process, we ensure the reproducibility of the results and the usability of the benchmark by the community. Those results serve as a basis of comparison for future studies. In a field where data are hard to obtain, and where the comparison of results from different studies is often irrelevant, GLYFE gives the opportunity of gathering researchers around a standardized common environment. [ABSTRACT FROM AUTHOR]

The most well-known and widely used mathematical representations of the physiology of a diabetic individual are the Sorensen and Hovorka models as well as the UVAPadova Simulator. While the Hovorka model and the UVAPadova Simulator only describe the glucose metabolism of a subject with type 1 diabetes, the Sorensen model was formulated to simulate the behaviour of both normal and diabetic individuals. The UVAPadova model is the most known model, accepted by the FDA, with a high level of complexity. The Hovorka model is the simplest of the three models, well documented and used primarily for the development of control algorithms. The Sorensen model is the most complete, even though some modifications were required both to the model equations (adding useful compartments for modelling subcutaneous insulin delivery) and to the parameter values. In the present work several simulated experiments, such as IVGTTs and OGTTs, were used as tools to compare the three formulations in order to establish to what extent increasing complexity translates into richer and more correct physiological behaviour. All the equations and parameters used for carrying out the simulations are provided. [ABSTRACT FROM AUTHOR]

Objective: To determine if a clinical presentation indistinguishable from migraine can occur due to an underlying condition or pathology, that is, "symptomatic migraine." Background: It is currently not clear whether migraine truly can be caused by an underlying condition or pathology. Characterization of the etiology and clinical features of possible symptomatic migraine is of significant clinical importance and further may help elucidate the pathophysiology of migraine. Methods: We devised operational diagnostic criteria for "symptomatic migraine" and "possible symptomatic migraine" requiring strong evidence for a causal relation between underlying cause and migraine symptoms adhering strictly to diagnostic criteria. PubMed was searched for case reports of symptomatic migraine from inception to March 2020. Only articles published in English or German were included. No restrictions were placed on study design. Relevant references in the articles were also included. Papers were systematically reviewed by two independent reviewers for detailed clinical features of migraine as well as the proposed underlying conditions and the effects of treatment of these conditions. Results: Our search retrieved 1726 items. After screening, 109 papers comprising 504 cases were reviewed in detail. Eleven patients with migraine with aura (MWA) fulfilled our working criteria for symptomatic migraine, and 39 patients fulfilled our criteria for possible symptomatic migraine. The most common etiologies of symptomatic migraine were arteriovenous malformations, carotid stenosis, dissection or aneurysm, brain infarctions, meningioma, and various intra‐axial tumors. Conclusions: Symptomatic MWA, indistinguishable from idiopathic MWA, may occur due to cortical lesions or microembolization. We found no clear evidence supporting the existence of symptomatic migraine without aura although we did identify possible cases. Our findings are limited by the available literature, and we suggest that prospective studies are needed. [ABSTRACT FROM AUTHOR]

In the thirty plus years since the Americans with Disabilities Act was passed, there have been a significant number of lengthy and costly judicial disputes involving medical school admission and enrollment of individuals with disabilities. This article reviews the history of medical education and provides a description of the evolution of the educational curriculum for medical school and how it has changed in recent years. It provides the legal framework of statutory and regulatory requirements for the application of federal disability discrimination law to medical school applicants and enrolled students. A synthesis of these cases (many lasting several years from incident to resolution) sheds light on what must be done, what can be done, and what should be done by medical school policy makers and administrators in response to the admission and enrollment of individuals with disabilities. The article suggests ways that medical schools could revise their evaluation procedures and practices both at the admissions stage and during medical school. The article stresses the importance of key top medical school leadership and medical school legal counsel in ensuring that this framework is implemented. The primary audience for this article are top administrators and legal counsel in institutions that set these policies and implement them. [ABSTRACT FROM AUTHOR]

This paper presents a mathematical model of glucose-insulin dynamics which is specific for type 2 diabetic patients. The general modelling is obtained by simplification of a global compartmental model by John Thomas Sorensen. The model parameters are estimated using nonlinear optimization and data collected in Rwanda for type 2 diabetic patients. In order to identify and evaluate possible abnormalities of type 2 diabetic patients, the Sampling Importance Resampling (SIR) particle filtering algorithm is used and implemented through discretization of the developed mathematical model. This process is done by clamping insulin and glucose concentrations at around clinical trial values as proposed by Defronzo. Furthermore, for detecting potential abnormalities in type 2 diabetic patients, we compare our results with results obtained from a simulation of the mathematical model for healthy subjects. The proposed mathematical model allows further investigation of the dynamic behavior of glucose, insulin, glucagon, stored insulin, and labile insulin in different organs for type 2 diabetic patients. [ABSTRACT FROM AUTHOR]

Mathematical modeling in the field of glucose metabolism has a longstanding tradition. The use of models is motivated by several reasons. Models have been used for calculating parameters of physiological interest from experimental data indirectly, to provide an unambiguous quantitative representation of pathophysiological mechanisms, to determine indices of clinical usefulness from simple experimental tests. With the growing societal impact of type 2 diabetes, which involves the disturbance of the glucose homeostasis system, development and use of models in this area have increased. Following the approaches of physiological and clinical investigation, the focus of the models has spanned from representations of whole body processes to those of cells, i.e., from in vivo to in vitro research. Model-based approaches for linking in vivo to in vitro research have been proposed, as well as multiscale models merging the two areas. The success and impact of models has been variable. Two kinds of models have received remarkable interest: those widely used in clinical applications, e.g., for the assessment of insulin sensitivity and β-cell function and some models representing specific aspects of the glucose homeostasis system, which have become iconic for their efficacy in describing clearly and compactly key physiological processes, such as insulin secretion from the pancreatic β cells. Models are inevitably simplified and approximate representations of a physiological system. Key to their success is an appropriate balance between adherence to reality, comprehensibility, interpretative value and practical usefulness. This has been achieved with a variety of approaches. Although many models concerning the glucose homeostasis system have been proposed, research in this area still needs to address numerous issues and tackle new opportunities. The mathematical representation of the glucose homeostasis processes is only partial, also because some mechanisms are still only partially understood. For in vitro research, mathematical models still need to develop their potential. This review illustrates the problems, approaches and contribution of mathematical modeling to the physiological and clinical investigation of glucose homeostasis and diabetes, focusing on the most relevant and stimulating models. [ABSTRACT FROM AUTHOR]

Background. Since physical activity has a high impact on patients with type 1 diabetes and the risk of hypoglycemia (low blood glucose levels) is significantly higher during and after physical activities, an automatic method to provide a personalized recommendation is needed to improve the blood glucose management and harness the benefits of physical activities. This paper aims to reduce the risk of hypoglycemia and hyperglycemia (high blood glucose levels), and empowers type 1 diabetes patients to make decisions regarding food choices connected with physical activities. Methods. Traditional and Bayesian feedforward neural network models are developed to provide accurate predictions of the blood glucose outcome and the risks of hyperglycemia and hypoglycemia with uncertainty information. Using the proposed models, safe actions that minimize the risk of both hypoglycemia and hyperglycemia are provided as food recommendations to the patient. Results. The predicted blood glucose responses to the optimal and safe food recommendations are significantly better and safer than by taking random food. Conclusions. Simulations conducted on the state-of-the-art UVA/Padova simulator combined with Brenton's physical activity model show that the proposed methodology is safe and effective in managing blood glucose during and after physical activities. [ABSTRACT FROM AUTHOR]