Blood Glucose Prediction Research Articles

EMD-Based Noninvasive Blood Glucose Estimation from PPG Signals Using Machine Learning Algorithms

Effective management of diabetes requires accurate monitoring of blood glucose levels. Traditional invasive methods for such monitoring can be cumbersome and uncomfortable for patients. In this study, we introduce a noninvasive approach to estimate blood glucose levels using photoplethysmography (PPG) signals. We have focused on blood glucose prediction using wrist PPG signals and explored various PPG waveform-based features, including AC to DC ratio (AC/DC) and intrinsic mode function (IMF)-based features derived from empirical mode decomposition (EMD). To the best of our knowledge, no studies have been found using EMD-based features to estimate blood glucose levels noninvasively. Additionally, feature importance-based selection has also been used to further improve the accuracy of the proposed model. Among the four machine learning algorithms considered in this study, CatBoost consistently outperformed XGBoost, LightGBM, and random forest across a wide number of features. The best performing model, CatBoost, achieved Pearson’s r of 0.96, MSE 0.08, R2 score 0.92, and MAE 8.01 when considering the top 50 features selected from both PPG waveform-based features and IMF-based features. The p-values for all models were <0.001, indicating statistically significant correlations. Overall, this study provides valuable insights into the feasibility and effectiveness of noninvasive blood glucose monitoring using advanced machine learning techniques.

An LSTM-based Neural Network Wearable System for Blood Glucose Prediction in People with Diabetes.

This article proposes the first hardware implemen-tation of a low-power LSTM neural network targeting a wearable medical device designed to predict blood glucose at a 30-minute horizon. This work aims to reduce energy consumption by propos-ing new activation functions that target hardware implementation. On top of this proposal, we also prove there is room for improve-ment in energy consumption by applying neural network optimiza-tions at the algorithmic, such as quantization, and architecture level, LSTM hyperparameters, that consider the target hardware. To validate our proposal, we devise an optimized version of the neural network aimed to be wearable and, therefore, to reduce its energy consumption while preserving its accuracy as much as possible. The hardware is implemented on a Xilinx Virtex-7 FPGA VC707 Evaluation Kit. It is compared with (i) a faithful design of the original neural network implemented on the same evaluation kit, (ii) three state-of-the-art LSTM-based FPGA implementations, and (iii) software implementations running in cutting-edge smartphones:OnePlus NordTM and an Apple iPhone 13 ProTM with artificial in-telligence hardware accelerators. Our proposal consumes between ×1020 and ×7 less energy than the software implementations, being the most efficient system compared to the smartphones. On the other hand, its energy efficiency, measured in GFLOP/J, is between ×2.84 and ×7.82 greater than other state-of-the-art LSTM implementations, proving to be the most suitable implementation for a wearable system for blood glucose prediction.

Predicting blood glucose level using salivary glucose and other associated factors: A machine learning model selection and evaluation study

IntroductionThere is a need for designing non-invasive methods to predict blood glucose levels to ensure timely diagnosis of Diabetes Mellitus. Needle anxiety and bleeding disorders preclude many from undertaking blood tests. ObjectivesThe primary objective of this study was to assess if biomarkers like saliva can be used to estimate blood glucose levels. The second objective was to develop and evaluate Machine Learning (ML) models to predict blood glucose levels based on salivary glucose and associated features. An insight into the patient's features, which was important for predicting blood glucose levels, was also required. MethodsA cross-sectional study was conducted, and blood and saliva samples, along with patient-related data, were collected from healthy and diabetic patients. ML techniques were applied to the data to develop a tool for predicting blood glucose levels using patient features. The prediction intervals were computed, clinical accuracy was assessed, and important features for the prediction were identified. ResultsThe Random Forest Regressor Model, with features identified using the wrapper method, was selected as the best, with an average RMSE of 43.28. The prediction intervals were computed for point estimate, MAE = 23.821, and coverage was 100 percent, the clinical accuracy was compared with that of glucometers and continuous monitoring systems. All predicted values are in Zones A and B of the Clarke error grid, and the bias was 6.41. The most important feature for predicting blood glucose level is salivary glucose level, followed by known risk factors like Family History, BMI, etc. The study found that salivary glucose levels are insufficient to classify blood glucose levels as high or normal. ConclusionThe study concluded that salivary glucose with associated patient features could be a potential non-invasive biomarker for predicting blood glucose levels in patients. The developed ML model could be deployed in a device that inputs patient features, analyzes salivary glucose, and can monitor blood glucose levels in a non-invasive manner. Further research is needed to validate the findings of this study and develop a proof of concept.

Short-term glucose prediction in Type 1 Diabetes

The use of short-term prediction techniques with the purpose of anticipating hyper- or hypoglycemic episodes for patients with type 1 diabetes has gained a lot of interest in the past years. Both hypoglycemia and hyperglycemia have a major impact and considerable effects in the life of a person with diabetes. The prediction of abnormal values of the blood glucose is a vital step in improving the life quality of a diabetic patient. This paper focuses on developing models for predicting blood glucose levels, to warn type 1 diabetes patients about potential hyper- or hypoglycemic events. Our main objective is to help the daily management of people with type 1 diabetes and to improve their long-term health condition. We will compare four models: a regression model, recurrent neural networks, Long Short-Term Memory, and Gated Recurrent Unit.

Deep Learning-Based Glucose Prediction Models: A Guide for Practitioners and a Curated Dataset for Improved Diabetes Management.

Accurate short- and mid-term blood glucose predictions are crucial for patients with diabetes struggling to maintain healthy glucose levels, as well as for individuals at risk of developing the disease. Consequently, numerous efforts from the scientific community have focused on developing predictive models for glucose levels. This study harnesses physiological data collected from wearable sensors to construct a series of data-driven models based on deep learning approaches. We systematically compare these models to offer insights for practitioners and researchers venturing into glucose prediction using deep learning techniques. Key questions addressed in this work encompass the comparison of various deep learning architectures for this task, determining the optimal set of input variables for accurate glucose prediction, comparing population-wide, fine-tuned, and personalized models, and assessing the impact of an individual's data volume on model performance. Additionally, as part of our outcomes, we introduce a meticulously curated dataset inclusive of data from both healthy individuals and those with diabetes, recorded in free-living conditions. This dataset aims to foster research in this domain and facilitate equitable comparisons among researchers.

A real-time deep learning-based system for the blood glucose prediction of the diabetic patients

A real-time deep learning-based system for the blood glucose prediction of the diabetic patients

Enhancing Personalized Blood Glucose Prediction: Deep Learning Insights From Ablation Study on Wearable Sensor Data

Enhancing Personalized Blood Glucose Prediction: Deep Learning Insights From Ablation Study on Wearable Sensor Data

A Machine Learning Web App to Predict Diabetic Blood Glucose Based on a Basic Noninvasive Health Checkup, Sociodemographic Characteristics, and Dietary Information: Case Study.

Over the past few decades, diabetes has become a serious public health concern worldwide, particularly in Bangladesh. The advancement of artificial intelligence can be reaped in the prediction of blood glucose levels for better health management. However, the practical validity of machine learning (ML) techniques for predicting health parameters using data from low- and middle-income countries, such as Bangladesh, is very low. Specifically, Bangladesh lacks research using ML techniques to predict blood glucose levels based on basic noninvasive clinical measurements and dietary and sociodemographic information. To formulate strategies for public health planning and the control of diabetes, this study aimed to develop a personalized ML model that predicts the blood glucose level of urban corporate workers in Bangladesh. Based on the basic noninvasive health checkup test results, dietary information, and sociodemographic characteristics of 271 employees of the Bangladeshi Grameen Bank complex, 5 well-known ML models, namely, linear regression, boosted decision tree regression, neural network, decision forest regression, and Bayesian linear regression, were used to predict blood glucose levels. Continuous blood glucose data were used in this study to train the model, which then used the trained data to predict new blood glucose values. Boosted decision tree regression demonstrated the greatest predictive performance of all evaluated models (root mean squared error=2.30). This means that, on average, our model's predicted blood glucose level deviated from the actual blood glucose level by around 2.30 mg/dL. The mean blood glucose value of the population studied was 128.02 mg/dL (SD 56.92), indicating a borderline result for the majority of the samples (normal value: 140 mg/dL). This suggests that the individuals should be monitoring their blood glucose levels regularly. This ML-enabled web application for blood glucose prediction helps individuals to self-monitor their health condition. The application was developed with communities in remote areas of low- and middle-income countries, such as Bangladesh, in mind. These areas typically lack health facilities and have an insufficient number of qualified doctors and nurses. The web-based application is a simple, practical, and effective solution that can be adopted by the community. Use of the web application can save money on medical expenses, time, and health management expenses. The created system also aids in achieving the Sustainable Development Goals, particularly in ensuring that everyone in the community enjoys good health and well-being and lowering total morbidity and mortality.

Evaluating machine and deep learning techniques in predicting blood sugar levels within the E-health domain

This paper focuses on exploring and comparing different machine learning algorithms in the context of diabetes management. The aim is to understand their characteristics, mathematical foundations, and practical implications specifically for predicting blood glucose levels. The study provides an overview of the algorithms, with a particular emphasis on deep learning techniques such as Long Short-Term Memory Networks. Efficiency is a crucial factor in practical machine learning applications, especially in the context of diabetes management. Therefore, the paper investigates the trade-off between accuracy, resource utilisation, time consumption, and computational power requirements, aiming to identify the optimal balance. By analysing these algorithms, the research uncovers their distinct behaviours and highlights their dissimilarities, even when their analytical underpinnings may appear similar.

Individualized Models for Glucose Prediction in Type 1 Diabetes: Comparing Black-Box Approaches to a Physiological White-Box One.

Accurate blood glucose (BG) prediction are key in next-generation tools for type 1 diabetes (T1D) management, such as improved decision support systems and advanced closed-loop control. Glucose prediction algorithms commonly rely on black-box models. Large physiological models, successfully adopted for simulation, were little explored for glucose prediction, mostly because their parameters are hard to individualize. In this work, we develop a BG prediction algorithm based on a personalized physiological model inspired by the UVA/Padova T1D Simulator. Then we compare white-box and advanced black-box personalized prediction techniques. A personalized nonlinear physiological model is identified from patient data through a Bayesian approach based on Markov Chain Monte Carlo technique. The individualized model was integrated within a particle filter (PF) to predict future BG concentrations. The black-box methodologies considered are non-parametric models estimated via gaussian regression (NP), three deep learning methods: long-short-term-memory (LSTM), gated recurrent unit (GRU), temporal convolutional networks (TCN), and a recursive autoregressive with exogenous input model (rARX). BG forecasting performances are assessed for several prediction horizons (PH) on 12 individuals with T1D, monitored in free-living conditions under open-loop therapy for 10 weeks. NP models provide the most effective BG predictions by achieving a root mean square error (RMSE), RMSE = 18.99 mg/dL, RMSE = 25.72 mg/dL and RMSE = 31.60 mg/dL, significantly outperforming: LSTM, GRU (for PH = 30 minutes), TCN, rARX, and the proposed physiological model for PH=30, 45 and 60 minutes. Black-box strategies remain preferable for glucose prediction even when compared to a white-box model with sound physiological structure and individualized parameters.

A comparison of various deep learning models on blood glucose management of type 1 diabetes: A brief review

Type 1 diabetes has become one of the most common chronic diseases nowadays because patients pancreas cannot produce sufficient insulin, which helps the blood sugar to enter the cells, which will cause them to build up in the bloodstream and leads to complications and dis-eases. Therefore, a basal-bolus insulin therapy that contains daily insulin injections has be-come routine for patients with type 1 diabetes to help regulate the blood sugar level. To better monitor the blood glucose level, digital health monitors are becoming the trending method for type 1 diabetes patients. Meanwhile, all the data generated by the digital monitoring devices made researchers realize that deep learning algorithms could be implanted to help the device better predict a patients blood sugar levels. In this paper, we aim to present a review of testing several state-of-the-art deep-learning models on blood glucose prediction. We have identified a literature search and focused on the deep-learning algorithms for glucose management. After detailed explanations of each model, we employ them on one mutual dataset to identify the direct prediction results of each model and compare the pros and cons of these models accord-ing to the results report. While all these models have the most advanced frameworks, the lack of feature varieties and data availability becomes their limitations. However, followed by the increased focus on the digital health field, these challenges might soon get resolved, which leads to more comprehensive models that could be further deployed in clinical conditions.

Correlational analysis between salivary and blood glucose levels in individuals with and without diabetes mellitus: a cross-sectional study

Objective To estimate the association of patient-related demographic, socioeconomic status, physical activity, stress, and dietary factors influencing the relationship between salivary and blood glucose levels in individuals with and without diabetes mellitus (DM). Method This cross-sectional study was conducted on 166 participants with and without DM. Saliva and blood were collected to estimate the glucose levels. Age, gender, occupation, socioeconomic and education level, BMI, hip to waist circumference, stress, dietary pattern, lifestyle, physical activity, family history of diabetes, and type of diabetes were recorded. The association of saliva to predict blood glucose levels was analysed using Spearman Rank Correlation and how these patient-related factors influence the correlation was estimated for future machine learning models. The difference in medians for various groups was calculated using the Mann-Whitney U Test or Kruskal Wallis Test. Results Blood glucose level is not significantly correlated to salivary glucose level. However, a statistically significant difference in the median blood glucose levels for diabetic participants (median = 137) compared to healthy controls (p-value < .05) was noted. The correlation between blood and salivary glucose was more positive for higher levels of glucose (Spearman 0.4). Age, alcohol consumption, monthly wages, intake of vegetables, and socioeconomic status affect blood glucose levels. Conclusion A correlation between saliva and blood glucose levels in healthy individuals was weak. Saliva should only be used as a monitoring tool rather than a diagnostic tool and is more reliable for patients with poorly controlled diabetes mellitus.

The importance of interpreting machine learning models for blood glucose prediction in diabetes: an analysis using SHAP

Machine learning has become a popular tool for learning models of complex dynamics from biomedical data. In Type 1 Diabetes (T1D) management, these models are increasingly been integrated in decision support systems (DSS) to forecast glucose levels and provide preventive therapeutic suggestions, like corrective insulin boluses (CIB), accordingly. Typically, models are chosen based on their prediction accuracy. However, since patient safety is a concern in this application, the algorithm should also be physiologically sound and its outcome should be explainable. This paper aims to discuss the importance of using tools to interpret the output of black-box models in T1D management by presenting a case-of-study on the selection of the best prediction algorithm to integrate in a DSS for CIB suggestion. By retrospectively “replaying” real patient data, we show that two long-short term memory neural networks (LSTM) (named p-LSTM and np-LSTM) with similar prediction accuracy could lead to different therapeutic decisions. An analysis with SHAP—a tool for explaining black-box models’ output—unambiguously shows that only p-LSTM learnt the physiological relationship between inputs and glucose prediction, and should therefore be preferred. This is verified by showing that, when embedded in the DSS, only p-LSTM can improve patients’ glycemic control.

A Probabilistic Approach to Blood Glucose Prediction in Type 1 Diabetes Under Meal Uncertainties.

Currently, most reliable and commercialized artificial pancreas systems for type 1 diabetes are hybrid closed-loop systems, which require the user to announce every meal and its size. However, estimating the amount of carbohydrates in a meal and announcing each and every meal is an error-prone process that introduces important uncertainties to the problem, which when not considered, lead to sub-optimal outcomes of the controller. To address this problem, we propose a novel deep-learning-based model for probabilistic glucose prediction, called the Input and State Recurrent Kalman Network (ISRKN), which consists in the incorporation of an input and state Kalman filter in the latent space of a deep neural network so that the posterior distributions can be computed in closed form and the uncertainty can be propagated using the Kalman equations. In addition, the proposed architecture allows explicit estimation of the meal uncertainty distribution, whose parameters are encoded in the filter parameters. Results using the UVA/Padova simulator and data from a clinical trial show that the proposed model outperforms other probabilistic models using several probabilistic metrics across different degrees of distributional shifts.

Forecasting glycaemia for type 1 diabetes mellitus patients by means of IoMT devices

The chronic metabolic condition, Type 1 diabetes mellitus (DM1), is marked by consistent hyperglycemia due to the body's inability to produce sufficient insulin. This necessitates the patient's daily monitoring of blood glucose fluctuations to discern a trend and predict future glycemia, subsequently dictating the amount of external insulin needed to regulate glycemia effectively. However, this technique often grapples with a degree of inaccuracy, presenting potential hazards. Nonetheless, contemporary advancements in information and communication technologies (ICT) coupled with novel biological signal sensors offer a refreshing perspective for DM1 management by enabling comprehensive, continual patient health evaluation. Herein, burgeoning technological disruptions such as Big Data, the internet of medical things (IoMT), cloud computing, and machine learning algorithms (ML) could serve pivotal roles in the effective control of DM1. This paper delves into the exploration of the latest IoMT-based methodologies for the unbroken surveillance of DM1 management, facilitating a profound characterization of diabetic patients. The fusion of wearable technologies with machine learning strategies has the potential to yield robust models for short-term blood glucose prediction. The ambition of this study is to develop precise, individual-centric prediction models harnessing an array of pertinent factors. The study applied modeling techniques to a comprehensive dataset comprising glycaemia-associated biological attributes, sourced from an expansive passive monitoring campaign involving 40 DM1 patients. Leveraging the Random Forest method, the resulting models can predict glucose levels over a 30-min time span with an average error as minimal as 18.60 mg/dL for six-hour data and 26.21 mg/dL for a 45-minute prediction horizon, offering also a good performance in the prediction delay.

A Stacked Long Short-Term Memory Approach for Predictive Blood Glucose Monitoring in Women with Gestational Diabetes Mellitus.

Gestational diabetes mellitus (GDM) is a subtype of diabetes that develops during pregnancy. Managing blood glucose (BG) within the healthy physiological range can reduce clinical complications for women with gestational diabetes. The objectives of this study are to (1) develop benchmark glucose prediction models with long short-term memory (LSTM) recurrent neural network models using time-series data collected from the GDm-Health platform, (2) compare the prediction accuracy with published results, and (3) suggest an optimized clinical review schedule with the potential to reduce the overall number of blood tests for mothers with stable and within-range glucose measurements. A total of 190,396 BG readings from 1110 patients were used for model development, validation and testing under three different prediction schemes: 7 days of BG readings to predict the next 7 or 14 days and 14 days to predict 14 days. Our results show that the optimized BG schedule based on a 7-day observational window to predict the BG of the next 14 days achieved the accuracies of the root mean square error (RMSE) = 0.958 ± 0.007, 0.876 ± 0.003, 0.898 ± 0.003, 0.622 ± 0.003, 0.814 ± 0.009 and 0.845 ± 0.005 for the after-breakfast, after-lunch, after-dinner, before-breakfast, before-lunch and before-dinner predictions, respectively. This is the first machine learning study that suggested an optimized blood glucose monitoring frequency, which is 7 days to monitor the next 14 days based on the accuracy of blood glucose prediction. Moreover, the accuracy of our proposed model based on the fingerstick blood glucose test is on par with the prediction accuracies compared with the benchmark performance of one-hour prediction models using continuous glucose monitoring (CGM) readings. In conclusion, the stacked LSTM model is a promising approach for capturing the patterns in time-series data, resulting in accurate predictions of BG levels. Using a deep learning model with routine fingerstick glucose collection is a promising, predictable and low-cost solution for BG monitoring for women with gestational diabetes.

High-precision prediction of blood glucose concentration utilizing Fourier transform Raman spectroscopy and an ensemble machine learning algorithm

Raman spectroscopy has gained popularity in analyzing blood glucose levels due to its non-invasive identification and minimal interference from water. However, the challenge lies in how to accurately predict blood glucose concentrations in human blood using Raman spectroscopy. This paper researches a novel integrated machine learning algorithm called Bagging-ABC-ELM. The optimal input weights and biases of extreme learning machine (ELM) model are obtained by artificial bee colony (ABC) algorithm. The bagging algorithm is used to obtain a better the stability of the model and higher performance than ELM algorithm. The results show that the mean value of coefficient of determination is 0.9928, and root mean square error is 0.1928. Compared to other regression models, the Bagging-ABC-ELM model exhibited superior prediction accuracy, robustness, and generalization capability. The Bagging-ABC-ELM model presents a promising alternative for analyzing blood glucose levels in clinical and research settings.

Heterogeneous temporal representation for diabetic blood glucose prediction.

Background and aims: Blood glucose prediction (BGP) has increasingly been adopted for personalized monitoring of blood glucose levels in diabetic patients, providing valuable support for physicians in diagnosis and treatment planning. Despite the remarkable success achieved, applying BGP in multi-patient scenarios remains problematic, largely due to the inherent heterogeneity and uncertain nature of continuous glucose monitoring (CGM) data obtained from diverse patient profiles. Methodology: This study proposes the first graph-based Heterogeneous Temporal Representation (HETER) network for multi-patient Blood Glucose Prediction (BGP). Specifically, HETER employs a flexible subsequence repetition method (SSR) to align the heterogeneous input samples, in contrast to the traditional padding or truncation methods. Then, the relationships between multiple samples are constructed as a graph and learned by HETER to capture global temporal characteristics. Moreover, to address the limitations of conventional graph neural networks in capturing local temporal dependencies and providing linear representations, HETER incorporates both a temporally-enhanced mechanism and a linear residual fusion into its architecture. Results: Comprehensive experiments were conducted to validate the proposed method using real-world data from 112 patients in two hospitals, comparing it with five well-known baseline methods. The experimental results verify the robustness and accuracy of the proposed HETER, which achieves the maximal improvement of 31.42%, 27.18%, and 34.85% in terms of MAE, MAPE, and RMSE, respectively, over the second-best comparable method. Discussions: HETER integrates global and local temporal information from multi-patient samples to alleviate the impact of heterogeneity and uncertainty. This method can also be extended to other clinical tasks, thereby facilitating efficient and accurate capture of crucial pattern information in structured medical data.

A State-by-State online transfer learning framework with incremental clustering for blood glucose prediction

This study proposes a State-by-State online transfer learning framework for blood glucose (BG) prediction. We assume that the BG data can be divided into states, representing different states of the patient. A thoroughly reliable blood glucose prediction model should not only be able to mimic the patient's physiology, such as fasting, food intake, and exercise, but also cope with external disturbances, such as noise, exercise, unannounced meals, and stress, among other factors. The state of the fixed-length BG data is separated using a clustering method. For each state, an appropriate prediction model is trained. This framework transforms the online prediction of BG into the process of matching the prediction model online, using incremental clustering methods and online transfer learning methods to cope when there is no matching model for the new state data input. The method uses incremental clustering to accurately divide the state of patients' blood glucose time series, builds a prediction model by state, and uses online transfer learning to train a new model when data is insufficient. In conclusion, we obtained the highest prediction accuracy on the dataset using this strategy.

A 2-dimensional model framework for blood glucose prediction based on iterative learning control architecture.

The accurate, timely, and personalized prediction for future blood glucose (BG) levels is undoubtedly needed for further advancement of diabetes management technologies. Human inherent circadian rhythm and regular lifestyle resulting in similarity of daily glycemic dynamics play a positive role in the prediction of blood glucose. Inspired by the iterative learning control (ILC) method in the field of automatic control, a 2-dimensional (2-D) model framework is constructed to predict the future blood glucose levels by taking both the short-range information within a day (intra-day) and long-range information between days (inter-day) into account. In this framework, the radial basis function neural network was applied to capture nonlinear relationships in glycemic metabolism, that is, short-range temporal dependence and long-range contemporaneous dependence on previous days. We build models for each patient, and the models were tested on the in silico datasets at various prediction horizons (PHs). The learning model developed in the 2-D framework successfully increases the accuracy and reduces the delay of predictions. This modeling framework provides a new point of view for BG level prediction and contributes to the development of personalized glucose management, such as hypoglycemia warning and glycemic control.