Data mining for blood glucose prediction and knowledge discovery in diabetic patients: the METABO diabetes modeling and management system

Dynamic circadian fluctuations of glycemia in patients with type 2 diabetes mellitus

BACKGROUND

Diabetes mellitus (DM) has glucose variability that is of such relevance that the appearance of vascular complications in patients with DM has been attributed to hyperglycemic and dysglycemic events. It is known that T1D patients mainly have glycemic variability with a specific oscillatory pattern with specific circadian characteristics for each patient. However, it has not yet been determined whether an oscillation pattern represents the variability of glycemic in T2D. This is why our objective is to determine the characteristics of glycemic oscillations in T2D and generate a robust predictive model.

RESULTS

Showed that glycosylated hemoglobin, glycemia, and body mass index were all higher in patients with T2D than in controls (all p < 0.05). In addition, time in hyperglycemia and euglycemia was markedly higher and lower in the T2D group (p < 0.05), without significant differences for time in hypoglycemia. Standard deviation, coefficient of variation, and total power of glycemia were significantly higher in the T2D group than Control group (all p < 0.05). The oscillatory patterns were significantly different between groups (p = 0.032): the control group was mainly distributed at 2-3 and 6 days, whereas the T2D group showed a more homogeneous distribution across 2-3-to-6 days.

CONCLUSIONS

The predictive model of glycemia showed that it is possible to accurately predict hyper- and hypoglycemia events. Thus, T2D patients exhibit specific oscillatory patterns of glycemic control, which are possible to predict. These findings may help to improve the treatment of DM by considering the individual oscillatory patterns of patients.

Design and Development of Diabetes Management System Using Machine Learning

This paper describes the design and implementation of a software system to improve the management of diabetes using a machine learning approach and to demonstrate and evaluate its effectiveness in controlling diabetes. The proposed approach for this management system handles the various factors that affect the health of people with diabetes by combining multiple artificial intelligence algorithms. The proposed framework factors the diabetes management problem into subgoals: building a Tensorflow neural network model for food classification; thus, it allows users to upload an image to determine if a meal is recommended for consumption; implementing K-Nearest Neighbour (KNN) algorithm to recommend meals; using cognitive sciences to build a diabetes question and answer chatbot; tracking user activity, user geolocation, and generating pdfs of logged blood sugar readings. The food recognition model was evaluated with cross-entropy metrics that support validation using Neural networks with a backpropagation algorithm. The model learned features of the images fed from local Ghanaian dishes with specific nutritional value and essence in managing diabetics and provided accurate image classification with given labels and corresponding accuracy. The model achieved specified goals by predicting with high accuracy, labels of new images. The food recognition and classification model achieved over 95% accuracy levels for specific calorie intakes. The performance of the meal recommender model and question and answer chatbot was tested with a designed cross-platform user-friendly interface using Cordova and Ionic Frameworks for software development for both mobile and web applications. The system recommended meals to meet the calorific needs of users successfully using KNN (with k = 5) and answered questions asked in a human-like way. The implemented system would solve the problem of managing activity, dieting recommendations, and medication notification of diabetics.

Short-term prediction of glucose in type 1 diabetes using kernel adaptive filters

This study aims at presenting a nonlinear, recursive, multivariate prediction model of the subcutaneous glucose concentration in type 1 diabetes. Nonlinear regression is performed in a reproducing kernel Hilbert space, by either the fixed budget quantized kernel least mean square (QKLMS-FB) or the approximate linear dependency kernel recursive least-squares (KRLS-ALD) algorithm, such that a sparse model structure is accomplished. A multivariate feature set (i.e., subcutaneous glucose, food carbohydrates, insulin regime and physical activity) is used and its influence on short-term glucose prediction is investigated. The method is evaluated using data from 15 patients with type 1 diabetes in free-living conditions. In the case when all the input variables are considered: (i) the average root mean squared error (RMSE) of QKLMS-FB increases from 13.1 mg dL^-1 (mean absolute percentage error (MAPE) 6.6%) for a 15-min prediction horizon (PH) to 37.7 mg dL^-1 (MAPE 20.8%) for a 60-min PH and (ii) the RMSE of KRLS-ALD, being predictably lower, increases from 10.5 mg dL^-1 (MAPE 5.2%) for a 15-min PH to 31.8 mg dL^-1 (MAPE 18.0%) for a 60-min PH. Multivariate data improve systematically both the regularity and the time lag of the predictions, reducing the errors in critical glucose value regions for a PH ≥ 30 min. Graphical abstract ᅟ.

Non-linear dynamic modeling of glucose in type 1 diabetes with kernel adaptive filters

We propose a non-linear recursive solution to the problem of short-term prediction of glucose in type 1 diabetes. The Fixed Budget Quantized Kernel Least Mean Square (QKLMS-FB) algorithm is employed to construct a univariate model of subcutaneous glucose concentration, which: (i) handles nonlinearities by transforming the input space into a high-dimensional Reproducing Kernel Hilbert Space and, (ii) finds a sparse solution by retaining a representative subset of the training input vectors. The dataset comes from the continuous multi-day recordings of 15 type 1 patients in free-living conditions. QKLMS-FB produces an average root mean squared error of 18.66±3.19 mg/dl for a prediction horizon of 30 min with 82.04% of hypoglycemic readings and 93.30% of hyperglycemic ones being classified as clinically accurate or with benign errors. The effect of the prediction horizon is more evident in the hypoglycemic range.

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.

A Systematic Review for Mobile Monitoring Solutions in M-Health

A systematic review allows us to identify, assess, and interpret all possible relevant work associated with a question in particular or the subject of an area. Different authors can use several methodologies to learn about research related to their own research in different fields. The main objective of this review is to identify work, research and publications made in the field of the mobile monitoring of patients through some application or commercial or non-commercial solutions in m-Health. Next, we compare the different solutions with our solution, MoMo (Mobile Monitoring) Framework. MoMo is a solution that allows for patient mobile monitoring through mobile phones and biometric devices (blood pressure meter, glucometer and others). Our systematic review is based on the methodology of B. Kitchenham. She proposed specific guidelines for carrying out a systematic review in software engineering. We prepare our systematic review base in the selection of primary and secondary research related to mobile monitoring solutions following criteria with a specific weight to compare with each part of our research.

Mobile Monitoring Framework to Design Parameterized and Personalized m-Health Applications According to the Patient's Diseases

The development of personalized mobile monitoring applications is a complex work. Currently, the most of applications for patients monitoring through mobile devices, is not developed considering the particular characteristics of each patient, but these applications have been developed taking into account a general behavior depending on the diseases instead of the own patients. The diseases manifest different symptoms depending on the patient situation. Mary and John (hypothetic patients) have diabetes, but the same measurement of glucose for each one affects their health in a different way. This paper describes a framework that allows the development of mobile applications, personalized for each patient, in such a way that even if they have the same disease, the application will respond to the individual needs of each patient.

A Clinical Support System Based on Quality of Life Estimation

Quality of life is a concept influenced by social, economic, psychological, spiritual or medical state factors. More specifically, the perceived quality of an individual's daily life is an assessment of their well-being or lack of it. In this context, information technologies may help on the management of services for healthcare of chronic patients such as estimating the patient quality of life and helping the medical staff to take appropriate measures to increase each patient quality of life. This paper describes a Quality of Life estimation system developed using information technologies and the application of data mining algorithms to access the information of clinical data of patients with cancer from Otorhinolaryngology and Head and Neck services of an oncology institution. The system was evaluated with a sample composed of 3013 patients. The results achieved show that there are variables that may be significant predictors for the Quality of Life of the patient: years of smoking (p value 0.049) and size of the tumor (p value < 0.001). In order to assign the variables to the classification of the quality of life the best accuracy was obtained by applying the John Platt's sequential minimal optimization algorithm for training a support vector classifier. In conclusion data mining techniques allow having access to patients additional information helping the physicians to be able to know the quality of life and produce a well-informed clinical decision.

Comparative assessment of glucose prediction models for patients with type 1 diabetes mellitus applying sensors for glucose and physical activity monitoring

The present work presents the comparative assessment of four glucose prediction models for patients with type 1 diabetes mellitus (T1DM) using data from sensors monitoring blood glucose concentration. The four models are based on a feedforward neural network (FNN), a self-organizing map (SOM), a neuro-fuzzy network with wavelets as activation functions (WFNN), and a linear regression model (LRM), respectively. For the development and evaluation of the models, data from 10 patients with T1DM for a 6-day observation period have been used. The models' predictive performance is evaluated considering a 30-, 60- and 120-min prediction horizon, using both mathematical and clinical criteria. Furthermore, the addition of input data from sensors monitoring physical activity is considered and its effect on the models' predictive performance is investigated. The continuous glucose-error grid analysis indicates that the models' predictive performance benefits mainly in the hypoglycemic range when additional information related to physical activity is fed into the models. The obtained results demonstrate the superiority of SOM over FNN, WFNN, and LRM with SOM leading to better predictive performance in terms of both mathematical and clinical evaluation criteria.

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.

Diabetes self-management arrangements in Europe: a realist review to facilitate a project implemented in six countries

Background

Self-management of long term conditions can promote quality of life whilst delivering benefits to the financing of health care systems. However, rarely are the meso-level influences, likely to be of direct relevance to these desired outcomes, systematically explored. No specific international guidelines exist suggesting the features of the most appropriate structure and organisation of health care systems within which to situate self-management approaches and practices. This review aimed to identify the quantitative literature with regard to diabetes self-management arrangements currently in place within the health care systems of six countries (The United Kingdom, The Netherlands, Norway, Spain, Bulgaria, and Greece) and explore how these are integrated into the broader health care and welfare systems in each country.

Methods

The methodology for a realist review was followed. Publications of interest dating from 2000 to 2013 were identified through appropriate MeSH terms by a systematic search in six bibliographic databases. A search diary was maintained and the studies were assessed for their quality and risk of bias.

Results

Following the multi-step search strategy, 56 studies were included in the final review (the majority from the UK) reporting design methods and findings on 21 interventions and programmes for diabetes and chronic disease self-management. Most (11/21, 52%) of the interventions were designed to fit within the context of primary care. The majority (11/21, 52%) highlighted behavioural change as an important goal. Finally, some (5/21, 24%) referred explicitly to Internet-based tools.

Conclusions

This review is based on results which are derived from a total of at least 5,500 individuals residing in the six participating countries. It indicates a policy shift towards patient-centred self-management of diabetes in a primary care context. The professional role of diabetes specialist nurses, the need for multidisciplinary approaches and a focus on patient education emerge as fundamental principles in the design of relevant programmes. Socio-economic circumstances are relevant to the capacity to self-manage and suggest that any gains and progress will be hard to maintain during economic austerity. This realist review should be interpreted within the wider context of a whole systems approach regarding self-care support and chronic illness management.

Electronic supplementary material

The online version of this article (doi:10.1186/1472-6963-14-453) contains supplementary material, which is available to authorized users.

Closing the loop from continuous M-health monitoring to fuzzy logic-based optimized recommendations

Continuous sensing of health metrics might generate a massive amount of data. Generating clinically validated recommendations, out of these data, to patients under monitoring is of prime importance to protect them from risk of falling into severe health degradation. Physicians also can be supported with automated recommendations that gain from historical data and increasing learning cycles. In this paper, we propose a Fuzzy Expert System that relies on data collected from continuous monitoring. The monitoring scheme implements preprocessing of data for better data analytics. However, data analytics implements the loopback feature in order to constantly improve fuzzy rules, knowledge base, and generated recommendations. Both techniques reduced data quantity, improved data quality and proposed recommendations. We evaluate our solution through a series of experiments and the results we have obtained proved that our fuzzy expert system combined with the intelligent monitoring and analytic techniques provide a high accuracy of collected data and valid advices.

A predictive model of subcutaneous glucose concentration in type 1 diabetes based on Random Forests

In this study, an individualized predictive model of the subcutaneous glucose concentration in type 1 diabetes is presented, which relies on the Random Forests regression technique. A multivariate dataset is utilized concerning the s.c. glucose profile, the plasma insulin concentration, the intestinal absorption of meal-derived glucose and the daily energy expenditure. In an attempt to capture daily rhythms in glucose metabolism, we also introduce a time feature in the predictive analysis. The dataset comes from the continuous multi-day recordings of 27 type 1 patients in free-living conditions. Evaluating the performance of the proposed method by 10-fold cross validation, an average RMSE of 6.60, 8.15, 9.25 and 10.83 mg/dl for 15, 30, 60 and 120 min prediction horizons, respectively, was attained.

A glucose model based on support vector regression for the prediction of hypoglycemic events under free-living conditions

BACKGROUND

The prevention of hypoglycemic events is of paramount importance in the daily management of insulin-treated diabetes. The use of short-term prediction algorithms of the subcutaneous (s.c.) glucose concentration may contribute significantly toward this direction. The literature suggests that, although the recent glucose profile is a prominent predictor of hypoglycemia, the overall patient's context greatly impacts its accurate estimation. The objective of this study is to evaluate the performance of a support vector for regression (SVR) s.c. glucose method on hypoglycemia prediction.

MATERIALS AND METHODS

We extend our SVR model to predict separately the nocturnal events during sleep and the non-nocturnal (i.e., diurnal) ones over 30-min and 60-min horizons using information on recent glucose profile, meals, insulin intake, and physical activities for a hypoglycemic threshold of 70 mg/dL. We also introduce herein additional variables accounting for recurrent nocturnal hypoglycemia due to antecedent hypoglycemia, exercise, and sleep. SVR predictions are compared with those from two other machine learning techniques.

RESULTS

The method is assessed on a dataset of 15 patients with type 1 diabetes under free-living conditions. Nocturnal hypoglycemic events are predicted with 94% sensitivity for both horizons and with time lags of 5.43 min and 4.57 min, respectively. As concerns the diurnal events, when physical activities are not considered, the sensitivity is 92% and 96% for a 30-min and 60-min horizon, respectively, with both time lags being less than 5 min. However, when such information is introduced, the diurnal sensitivity decreases by 8% and 3%, respectively. Both nocturnal and diurnal predictions show a high (>90%) precision.

CONCLUSIONS

Results suggest that hypoglycemia prediction using SVR can be accurate and performs better in most diurnal and nocturnal cases compared with other techniques. It is advised that the problem of hypoglycemia prediction should be handled differently for nocturnal and diurnal periods as regards input variables and interpretation of results.

Multivariate prediction of subcutaneous glucose concentration in type 1 diabetes patients based on support vector regression

Data-driven techniques have recently drawn significant interest in the predictive modeling of subcutaneous (s.c.) glucose concentration in type 1 diabetes. In this study, the s.c. glucose prediction is treated as a multivariate regression problem, which is addressed using support vector regression (SVR). The proposed method is based on variables concerning: (i) the s.c. glucose profile, (ii) the plasma insulin concentration, (iii) the appearance of meal-derived glucose in the systemic circulation, and (iv) the energy expenditure during physical activities. Six cases corresponding to different combinations of the aforementioned variables are used to investigate the influence of the input on the daily glucose prediction. The proposed method is evaluated using a dataset of 27 patients in free-living conditions. 10-fold cross validation is applied to each dataset individually to both optimize and test the SVR model. In the case where all the input variables are considered, the average prediction errors are 5.21, 6.03, 7.14 and 7.62 mg/dl for 15, 30, 60 and 120 min prediction horizons, respectively. The results clearly indicate that the availability of multivariable data and their effective combination can significantly increase the accuracy of both short-term and long-term predictions.

Integration of heterogeneous biomedical sensors into an ISO/IEEE 11073 compliant application

Current trends in healthcare technology include mobile-based applications. Relevant advances in the integration of vital signs monitoring devices with mobile platforms are widely reported nowadays. In this context, conceiving and designing an interoperable application is essential due to the growing necessity of integrating a huge and heterogeneous amount of biomedical data, coming from a wide range of devices and sensors. In this paper the key research issues associated with such integration are presented as well as a specific proposal to solve these problems. It is based on a middleware architecture for the integration of biomedical sensors with mobile devices, derived from the ISO/IEEE 11073 standards family. The application has been developed in the framework of an EU-funded R&D project called METABO.

Predictive modeling of glucose metabolism using free-living data of type 1 diabetic patients

The aim of this study is to model the blood glucose metabolism of type 1 diabetic patients using free-living data. The proposed method considers the effect of diet, medication and exercise on blood glucose levels. Compartmental models are used to quantify the absorption of subcutaneously administered insulin, the absorption of glucose from the gut following a meal, as well as the effects of exercise on plasma glucose and insulin dynamics. Compartmental analysis is combined with a glucose predictive model which employs Support Vector machines for Regression to estimate the subcutaneous glucose concentrations. The model is trained and tested on real data recorded from two type 1 diabetic patients. The results obtained demonstrate the ability of the model to predict glucose response with a sufficient accuracy.