Blood Glucose Prediction Research Articles

A new multivariate blood glucose prediction method with hybrid feature clustering and online transfer learning

A new multivariate blood glucose prediction method with hybrid feature clustering and online transfer learning

Risk Prediction of high blood glucose among women (15–49 years) and men (15–54 years) in India: An analysis from National Family Health Survey-5 (2019–21)

ABSTRACT Context: Approximately 500 million individuals worldwide are known to have diabetes, representing roughly 1 out of every 11 adults in the world. Approximately 45.8% of adult diabetes cases are believed to be undiagnosed. Aim: This study aimed to identify the predictors for high blood glucose and to develop a risk score which helps in early detection of high blood glucose among Indian men (15–54 years) and women (15–49 years). Methods and Material: This study utilised data from the National Family Health Survey-5, which were gathered between 2019 and 2021. The study population comprises women aged 15–49 years and men aged 15–54 years in India. Statistical Analysis Used: A logistic regression analysis was conducted to determine the predictors of high blood glucose. The results were expressed as odds ratios with 95% confidence intervals. The risk score for high blood glucose was derived through variable shrinking and by employing regression coefficients obtained from the standard logistic regression model. Data were analysed using IBM SPSS version 26. Results: The prevalence of high blood glucose in India was 9.3%. The study findings indicated an association between age and the occurrence of high blood glucose levels. The prevalence of high blood glucose was higher among males (11.1% vs 7.5%), individuals living in urban areas (10.7% vs 8.9%), those with a waist circumference exceeding the specified limit (11.7% vs 5.9%), and individuals who were overweight or obese (11.3%). The prevalence of high blood glucose was higher among alcoholics (13.2% vs 8.8%) and various forms of tobacco users (12.1% vs 8.4%). Conclusions: Age, sex, place of residence (urban), consumption of alcohol, hypertension, and waist circumference were found to be the significant predictor variables and were used to develop the risk prediction score using the logistic regression model.

Significant of Blood Glucose Time-Series Forecasting with Temporal Fusion Transformer Model for Type 1 Diabetes

Blood glucose (BG) prediction has advanced to a state of the art thanks to deep learning models, which have been demonstrated to enhance type 1 diabetes (T1D) therapy. Nevertheless, most current models are limited to single-horizon prediction and have several practical drawbacks, including poor interpretability. In this study, we develop a novel approach to optimize the forecasting model in the training processes using the Optuna function, cross-validation, and the latest dataset. We offer a novel deep learning framework for multi-horizon BG prediction: the temporal fusion transformer (TFT). TFT employs a self-attention mechanism to extract long-term temporal dependencies and enables a model with an auto-tuning adjustment approach on hyperparameters and a cross-validation function on univariate and multivariate input models. On a clinical dataset with T1D subjects, namely ShanghaiT1DM (16 subjects), D1NAMO (9 subjects), and OhioT1DM (6 subjects) datasets, it achieved an average root mean square error (RMSE) of the three datasets used. The TFT model gave a lower error score than the baseline models of neural hierarchical interpolation for time series (N-HiTS)and long short-term memory (LSTM) with the values of 10.08+0.31 mg/dL and 12.34+0.62 mg/dL on the univariate input model and 9.18+1.21 mg/dL and 14.33+0.52 mg/dL on the multivariate input model for the prediction horizons of 30 and 60 minutes, respectively. This result explained that the TFT model was adequate for carrying out multi-horizon BG value-level forecasting and potentially deploying on-edge devices to improve clinical efforts to manage BG levels for T1D patients in real-time application.

Enhanced Diabetes Detection and Blood Glucose Prediction Using TinyML-Integrated E-Nose and Breath Analysis: A Novel Approach Combining Synthetic and Real-World Data

Diabetes mellitus, a chronic condition affecting millions worldwide, necessitates continuous monitoring of blood glucose level (BGL). The increasing prevalence of diabetes has driven the development of non-invasive methods, such as electronic noses (e-noses), for analyzing exhaled breath and detecting biomarkers in volatile organic compounds (VOCs). Effective machine learning models require extensive patient data to ensure accurate BGL predictions, but previous studies have been limited by small sample sizes. This study addresses this limitation by employing conditional generative adversarial networks (CTGAN) to generate synthetic data from real-world tests involving 29 healthy and 29 diabetic participants, resulting in over 14,000 new synthetic samples. These data were used to validate machine learning models for diabetes detection and BGL prediction, integrated into a Tiny Machine Learning (TinyML) e-nose system for real-time analysis. The proposed models achieved an 86% accuracy in BGL identification using LightGBM (Light Gradient Boosting Machine) and a 94.14% accuracy in diabetes detection using Random Forest. These results demonstrate the efficacy of enhancing machine learning models with both real and synthetic data, particularly in non-invasive systems integrating e-noses with TinyML. This study signifies a major advancement in non-invasive diabetes monitoring, underscoring the transformative potential of TinyML-powered e-nose systems in healthcare applications.

Breath Analysis Using Quartz Tuning Forks for Predicting Blood Glucose Levels Using Artificial Neural Networks.

Diabetes Mellitus (DM), a widespread metabolic disorder, poses lifelong health implications, demanding timely diagnosis and cautious monitoring for effective disease management. Traditional blood glucose tests are invasive and require medical expertise for intermittent checking, motivating the investigation of alternative, noninvasive methods. This study introduces an approach employing breath analysis through a set of 12 quartz tuning fork-based sensors enhanced using nanomaterials and dedicated artificial neural network (ANN) algorithms for data interpretation. The breath analysis methodology involves capturing unique breath signatures using the frequency-based sensor array. The accompanying neural network classification algorithm, customized for the sensor data, enables precise classification of data from 245 individuals as diabetic, prediabetic, or healthy. A neural network regression algorithm predicted blood glucose values and was compared with the actual values obtained from medical blood glucose measurement. The clinical relevance of the predicted blood glucose has been examined using error grids. The sensor array coupled with the ANN algorithm can identify diabetic, prediabetic, and control samples with 97% test accuracy. Blood glucose was predicted using neural network regression with a correlation coefficient of 0.89 and a mean square error of 0.13.

Predicting Neonatal Hypoglycemia Using AI Neural Networks in Infants from Mothers with Gestational Diabetes Mellitus.

BACKGROUND This study aimed to develop a predictive model for the association between maternal and neonatal anthropometric data and neonatal hypoglycemia based on data from mothers with gestational diabetes mellitus (GDM) and their neonates. MATERIAL AND METHODS We included 106 pregnant women with GDM (based on the World Health Organization International Association of Diabetes and Pregnancy Study Groups) and their neonates. Neonatal hypoglycemia was defined as a threshold of 2.5 mmol/L. Neonatal blood glucose levels were performed at 0, 0.5, 1, 3, and 24 h after birth. An artificial neural network (ANN) and recurrent neural network (RNN) were developed to predict the neonate blood concentrations and investigate the relative contribution of maternal and neonate clinical variables to neonatal hypoglycemia. RESULTS Of 106 mothers with GDM, 85% had obesity, and 78% had vaginal deliveries, with neonates averaging a birth weight of 3335.83 g. The ANN model, based on the clinical data from mothers and neonates, predicted blood glucose levels with a high degree of accuracy, achieving a coefficient of determination of 0.869 and a root mean square error (RMSE) of 0.274. Neonatal birth weight and maternal body mass index were the 2 most significant factors in predicting neonatal hypoglycemia, contributing 18.6% and 15.9%, respectively. The RNN model similarly forecasted glucose levels effectively, addressing the dynamic changes in blood glucose with 0.63 mmol/L RMSE and 0.53 mmol/L mean absolute error. CONCLUSIONS ANN and RNN models effectively predict neonatal hypoglycemia in infants of mothers with GDM, highlighting the critical role of maternal and neonatal factors.

Deep learning for blood glucose level prediction: How well do models generalize across different data sets?

Deep learning-based models for predicting blood glucose levels in diabetic patients can facilitate proactive measures to prevent critical events and are essential for closed-loop control therapy systems. However, selecting appropriate models from the literature may not always yield conclusive results, as the choice could be influenced by biases or misleading evaluations stemming from different methodologies, datasets, and preprocessing techniques. This study aims to compare and comprehensively analyze the performance of various deep learning models across diverse datasets to assess their applicability and generalizability across a broader spectrum of scenarios. Commonly used deep learning models for blood glucose level forecasting, such as feed-forward neural network, convolutional neural network, long short-term memory network (LSTM), temporal convolutional neural network, and self-attention network (SAN), are considered in this study. To evaluate the generalization capabilities of each model, four datasets of varying sizes, encompassing samples from different age groups and conditions, are utilized. Performance metrics include Root Mean Square Error (RMSE), Mean Absolute Difference (MAD), and Coefficient of Determination (CoD) for analytical asssessment, Clarke Error Grid (CEG) for clinical assessments, Kolmogorov-Smirnov (KS) test for statistical analysis, and generalization ability evaluations to obtain both coarse and granular insights. The experimental findings indicate that the LSTM model demonstrates superior performance with the lowest root mean square error and highest generalization capability among all other models, closely followed by SAN. The ability of LSTM and SAN to capture long-term dependencies in blood glucose data and their correlations with various influencing factors and events contribute to their enhanced performance. Despite the lower predictive performance, the FFN was able to capture patterns and trends in the data, suggesting its applicability in forecasting future direction. Moreover, this study helps in identifying the optimal model based on specific objectives, whether prioritizing generalization or accuracy.

Continuous glucose monitoring using machine learning models and IoT device data: A meta-analysis.

Machine learning offers diverse options for effectively managing blood glucose levels in diabetes patients. Selecting the right ML algorithm is critical given the array of available choices. Integrating data from IoT devices presents promising opportunities to enhance real-time blood glucose management models. This meta-analysis aims to evaluate the effectiveness of machine learning models utilizing IoT device data for predicting blood glucose levels. We systematically searched electronic databases for studies published between 2019 and 2023. We excluded studies lacking ML model derivation or performance metrics. The Quality Assessment of Diagnostic Accuracy Studies tool assessed study quality. Our primary outcomes compared ML models for BG level prediction across different prediction horizons (PHs). We analyzed ten eligible studies across prediction horizons of 15, 30, 45, and 60 minutes. ML models exhibited mean absolute RMSE values of 15.02 (SD 1.45), 21.488 (SD 2.92), 30.094 (SD 3.245), and 35.89 (SD 6.4) mg/dL, respectively. Random Forest demonstrated superior performance across these PHs. We observed significant heterogeneity across all subgroups, indicating diverse sources of variability. As the PH lengthened, the RMSE for blood glucose prediction by the ML model increased, with Random Forest showing the highest relative performance among the ML models.

BGformer: An improved Informer model to enhance blood glucose prediction

Accurately predicting blood glucose levels is crucial in diabetes management to mitigate patients’ risk of complications. However, blood glucose values exhibit instability, and existing prediction methods often struggle to capture their volatile nature, leading to inaccurate trend forecasts. To address these challenges, we propose a novel blood glucose level prediction model based on the Informer architecture: BGformer. Our model introduces a feature enhancement module and a microscale overlapping concerns mechanism. The feature enhancement module integrates periodic and trend feature extractors, enhancing the model’s ability to capture relevant information from the data. By extending the feature extraction capacity of time series data, it provides richer feature representations for analysis. Meanwhile, the microscale overlapping concerns mechanism adopts a window-based strategy, computing attention scores only within specific windows. This approach reduces computational complexity while enhancing the model’s capacity to capture local temporal dependencies. Furthermore, we introduce a dual attention enhancement module to augment the model’s expressive capability. Through prediction experiments on blood glucose values from sixteen diabetic patients, our model outperformed eight benchmark models in terms of both MAE and RMSE metrics for future 60-minute and 90-minute predictions. Our proposed scheme significantly improves the model’s dependency-capturing ability, resulting in more accurate blood glucose level predictions.

Blood Glucose Prediction from Nutrition Analytics in Type 1 Diabetes: A Review.

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. 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. 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%). 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.

Advancing elderly diabetes care: exploring the usability and acceptance of continuous glucose monitoring (CGM)

ObjectivesThis study aimed to evaluate the acceptability, usability, compliance, and satisfaction of Continuous Glucose Monitoring (CGM) among elderly diabetic patients. DesignUtilizing a mixed-methods approach, the study involved 30 participants, averaging 74.79 years in age. Participants wore two CGM devices, and their experiences, along with survey data, were comprehensively analyzed to gauge the effectiveness of CGM in managing diabetes. ResultsThe study demonstrated that -CGM- is highly usable and acceptable among elderly diabetic patients. Participants effectively utilized CGM to monitor and predict blood glucose trends, positively influencing their glucose control and lifestyle. The average adherence rate was found to be 81 %, indicating a substantial level of effective self-management and treatment decision-making. ConclusionThe study recommends developing CGM educational programs tailored for the elderly, educating healthcare professionals, expanding insurance coverage for CGM, and promoting real-time CGM technology to improve usability and acceptance among the elderly.

Comment on Martínez-Delgado et al. Using Absorption Models for Insulin and Carbohydrates and Deep Leaning to Improve Glucose Level Predictions. Sensors 2021, 21, 5273.

The paper "Using Absorption Models for Insulin and Carbohydrates and Deep Leaning to Improve Glucose Level Predictions" (Sensors2021, 21, 5273) proposes a novel approach to predicting blood glucose levels for people with type 1 diabetes mellitus (T1DM). By building exponential models from raw carbohydrate and insulin data to simulate the absorption in the body, the authors reported a reduction in their model's root-mean-square error (RMSE) from 15.5 mg/dL (raw) to 9.2 mg/dL (exponential) when predicting blood glucose levels one hour into the future. In this comment, we demonstrate that the experimental techniques used in that paper are flawed, which invalidates its results and conclusions. Specifically, after reviewing the authors' code, we found that the model validation scheme was malformed, namely, the training and test data from the same time intervals were mixed. This means that the reported RMSE numbers in the referenced paper did not accurately measure the predictive capabilities of the approaches that were presented. We repaired the measurement technique by appropriately isolating the training and test data, and we discovered that their models actually performed dramatically worse than was reported in the paper. In fact, the models presented in the that paper do not appear to perform any better than a naive model that predicts future glucose levels to be the same as the current ones.

Predicting Blood Glucose Levels with Organic Neuromorphic Micro-Networks.

Accurate glucose prediction is vital for diabetes management. Artificial intelligence and artificial neural networks (ANNs) are showing promising results for reliable glucose predictions, offering timely warnings for glucose fluctuations. The translation of these software-based ANNs into dedicated computing hardware opens a route toward automated insulin delivery systems ultimately enhancing the quality of life for diabetic patients. ANNs are transforming this field, potentially leading to implantable smart prediction devices and ultimately to a fully artificial pancreas. However, this transition presents several challenges, including the need for specialized, compact, lightweight, and low-power hardware. Organic polymer-based electronics are a promising solution as they have the ability to implement the behavior of neural networks, operate at low voltage, and possess key attributes like flexibility, stretchability, and biocompatibility. Here, the study focuses on implementing software-based neural networks for glucose prediction into hardware systems. How to minimize network requirements, downscale the architecture, and integrate the neural network with electrochemical neuromorphic organic devices, meeting the strict demands of smart implants for in-body computation of glucose prediction is investigated.

Optimizing type 2 diabetes management: AI-enhanced time series analysis of continuous glucose monitoring data for personalized dietary intervention.

Despite advanced health facilities in many developed countries, diabetic patients face multifold health challenges. Type 2 diabetes mellitus (T2DM) go along with conspicuous symptoms due to frequent peaks, hypoglycemia <=70 mg/dL (while fasting), or hyperglycemia >=180 mg/dL two hours postprandial, according to the American Diabetes Association (ADA)). The worse effects of Type 2 diabetes mellitus are precisely associated with the poor lifestyle adopted by patients. In particular, a healthy diet and nutritious food are the key to success for such patients. This study was done to help T2DM patients improve their health by developing a favorable lifestyle under an AI-assisted Continuous glucose monitoring (CGM) digital system. This study aims to reduce the blood glucose level fluctuations of such patients by rectifying their daily diet and maintaining their exertion vs. food consumption records. In this study, a well-precise prediction is obtained by training the ML model on a dataset recorded from CGM sensor devices attached to T2DM patients under observation. As the data obtained from the CGM sensor is time series, to predict blood glucose levels, the time series analysis and forecasting are done with XGBoost, SARIMA, and Prophet. The results of different Models are then compared based on performance metrics. This helped in monitoring various trends, specifically irregular patterns of the patient's glucose data, collected by the CGM sensor. Later, keeping track of these trends and seasonality, the diet is adjusted accordingly by adding or removing particular food and keeping track of its nutrients with the intervention of a commercially available all-in-one AI solution for food recognition. This created an interactive assistive system, where the predicted results are compared to food contents to bring the blood glucose levels within the normal range for maintaining a healthy lifestyle and to alert about blood glucose fluctuations before the time that are going to occur sooner. This study will help T2DM patients get in managing diabetes and ultimately bring HbA1c within the normal range (<= 5.7%) for diabetic and pre-diabetic patients, three months after the intervention.

Development of sensor system and data analytic framework for non-invasive blood glucose prediction

AbstractPeriodic quantification of blood glucose levels is performed using painful, invasive methods. The proposed work presents the development of a noninvasive glucose-monitoring device with two sensors, i.e., finger and wrist bands. The sensor system was designed with a near-infrared (NIR) wavelength of 940 nm emitter and a 900–1700 nm detector. This study included 101 diabetic and non-diabetic volunteers. The obtained dataset was subjected to pre-processing, exploratory data analysis (EDA), data visualization, and integration methods. Ambiguities such as the effects of skin color, ambient light, and finger pressure on the sensor were overcome in the proposed ‘niGLUC-2.0v’. niGLUC-2.0v was validated with performance metrics where accuracy of 99.02%, mean absolute error (MAE) of 0.15, mean square error (MSE) of 0.22 for finger, and accuracy of 99.96%, MAE of 0.06, MSE of 0.006 for wrist prototype with ridge regression (RR) were achieved. Bland–Altman analysis was performed, where 98% of the data points were within ± 1.96 standard deviation (SD), 100% were under zone A of the Clarke Error Grid (CEG), and statistical analysis showed p &lt; 0.05 on evaluated accuracy. Thus, niGLUC-2.0v is suitable in the medical and personal care fields for continuous real-time blood glucose monitoring.

Using nonlinear auto-regressive with exogenous input neural network (NNARX) in blood glucose prediction

BackgroundPredicting of future blood glucose (BG) concentration is important for diabetes control. Many automatic BG monitoring or controlling systems use BG predictors. The accuracy of the prediction for long prediction time is a major factor affecting the performance of the control system. The predicted BG can be used for glycemia management in the form of early hypoglycemic/hyperglycemic alarms or adjusting insulin injections. Recent developments in continuous glucose monitoring (CGM) devices open new opportunities for glycemia management of diabetic patients. Many of those systems need prediction for long prediction horizons to avoid going through hypo or hyperglycemia.MethodsIn this article a nonlinear autoregressive exogenous input neural network (NNARX) is proposed to predict the glucose concentration for longer prediction horizons (PHs) than that was obtained previously with an established recurrent neural network (RNN). The proposed NNARX is a modified version from our previously published RNN with different initialization and building technique but has the same architecture. The modification is based on starting with building nonlinear autoregressive exogenous input model using MATLAB and train it, then close the loop to get NNARX network.ResultsThe results of using the proposed NNARX indicate that the proposed NNARX is better in prediction and stability than unmodified RNN as PH becomes higher than 45 minutes.ConclusionsModification in RNN building extends the ability of the prediction till 100 minutes. It performs statistically significant improvements in the FIT and RMSE values for 100 minutes prediction. It also decreases root mean squared error (RMSE) for both 45 and 60 minutes of prediction.

Population-Specific Glucose Prediction in Diabetes Care With Transformer-Based Deep Learning on the Edge.

Leveraging continuous glucose monitoring (CGM) systems, real-time blood glucose (BG) forecasting is essential for proactive interventions, playing a crucial role in enhancing the management of type 1 diabetes (T1D) and type 2 diabetes (T2D). However, developing a model generalized to a population and subsequently embedding it within a microchip of a wearable device presents significant technical challenges. Furthermore, the domain of BG prediction in T2D remains under-explored in the literature. In light of this, we propose a population-specific BG prediction model, leveraging the capabilities of the temporal fusion Transformer (TFT) to adjust predictions based on personal demographic data. Then the trained model is embedded within a system-on-chip, integral to our low-power and low-cost customized wearable device. This device seamlessly communicates with CGM systems through Bluetooth and provides timely BG predictions using edge computing. When evaluated on two publicly available clinical datasets with a total of 124 participants with T1D or T2D, the embedded TFT model consistently demonstrated superior performance, achieving the lowest prediction errors when compared with a range of machine learning baseline methods. Executing the TFT model on our wearable device requires minimal memory and power consumption, enabling continuous decision support for more than 51 days on a single Li-Poly battery charge. These findings demonstrate the significant potential of the proposed TFT model and wearable device in enhancing the quality of life for people with diabetes and effectively addressing real-world challenges.

Glu-Ensemble: An ensemble deep learning framework for blood glucose forecasting in type 2 diabetes patients

Diabetes is a chronic metabolic disorder characterized by elevated blood glucose levels, posing significant health risks such as cardiovascular disease, and nerve, kidney, and eye damage. Effective management of blood glucose is essential for individuals with diabetes to mitigate these risks. This study introduces the Glu-Ensemble, a deep learning framework designed for precise blood glucose forecasting in patients with type 2 diabetes. Unlike other predictive models, Glu-Ensemble addresses challenges related to small sample sizes, data quality issues, reliance on strict statistical assumptions, and the complexity of models. It enhances prediction accuracy and model generalizability by utilizing larger datasets and reduces bias inherent in many predictive models. The framework's unified approach, as opposed to patient-specific models, eliminates the need for initial calibration time, facilitating immediate blood glucose predictions for new patients. The obtained results indicate that Glu-Ensemble surpasses traditional methods in accuracy, as measured by root mean square error, mean absolute error, and error grid analysis. The Glu-Ensemble framework emerges as a promising tool for blood glucose level prediction in type 2 diabetes patients, warranting further investigation in clinical settings for its practical application.

The effect of high ethanol concentration on E-nose response for diabetes detection in exhaled breath: Laboratory studies

The growing number of people with diabetes has paved the way for the creation of non-invasive devices to measure blood glucose levels and enable the detection of diabetes non-invasively, for example, through exhaled air. This paper presents the development of a breath detection system using multiple sensors (e-nose system) for the non-invasive estimation of blood glucose levels. The system employs commercially available ethanol, CO2, and acetone sensors, with synthetic breath mixtures used for testing, including four scenarios with high ethanol concentrations (0–570 ppm) as an influence factor. Results showed a mean absolute error of 0.245 ppm when analysing commonly observed acetone concentrations in diabetic breath using four sensors and the XGBoost Regressor. In mixtures with high ethanol concentrations and varying acetone concentrations (0–8.62 ppm), the CatBoost Regressor outperformed other machine learning algorithms with a mean absolute error of 0.568 ppm. This study emphasises the significant impact of ethanol on acetone detection and the need to consider ethanol levels in the developing of non-invasive devices for blood glucose prediction based on the exhaled acetone measurements. The research shows that a set of three gas sensors is optimal for estimating acetone concentrations in gas mixtures. The presented results constitute a preliminary step towards developing a non-invasive device for estimating blood glucose levels based on breath analysis, with the novelty of considering alcohol intake as a potential influencing factor.

AWD-stacking: An enhanced ensemble learning model for predicting glucose levels.

Accurate prediction of blood glucose levels is essential for type 1 diabetes optimizing insulin therapy and minimizing complications in patients with type 1 diabetes. Using ensemble learning algorithms is a promising approach. In this regard, this study proposes an improved stacking ensemble learning algorithm for predicting blood glucose level, in which three improved long short-term memory network models are used as the base model, and an improved nearest neighbor propagation clustering algorithm is adaptively weighted to this ensemble model. The OhioT1DM dataset is used to train and evaluate the performance of the proposed model. This study evaluated the performance of the proposed model using the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Matthews Correlation Coefficient (MCC) as the evaluation metrics. The experimental results demonstrate that the proposed model achieves an RMSE of 1.425 mg/dL, MAE of 0.721 mg/dL, and MCC of 0.982 mg/dL for a 30-minute prediction horizon(PH), RMSE of 3.212 mg/dL, MAE of 1.605 mg/dL, and MCC of 0.950 mg/dL for a 45-minute PH; and RMSE of 6.346 mg/dL, MAE of 3.232 mg/dL, and MCC of 0.930 mg/dL for a 60-minute PH. Compared with the best non-ensemble model StackLSTM, the RMSE and MAE were improved by up to 27.92% and 65.32%, respectively. Clarke Error Grid Analysis and critical difference diagram revealed that the model errors were within 10%. The model proposed in this study exhibits state-of-the-art predictive performance, making it suitable for clinical decision-making and of significant importance for the effective treatment of diabetes in patients.