Machine-learning Prediction of Hypo- and Hyperglycemia from Electronic Health Records: Algorithm Development and Validation

Predictive factors of hypoglycemia in type 2 diabetes: a prospective study using machine learning

Hypoglycemia is a serious complication in individuals with type 2 diabetes mellitus. Identifying who is most at risk remains challenging due to the non-linear relationships between hypoglycemia and its associated risk factors. The objective of this study is to evaluate the importance and impact of risk factors related to the incidence of hypoglycemia through an explainable machine learning method. This prospective study enrolled 1306 adults with type 2 diabetes mellitus at a specialized diabetes center. Over three months, participants were asked to do self-monitoring blood glucose measurements and record hypoglycemic events. Nine clinically relevant features were analyzed using five machine learning models. The performance of the models was evaluated by different metrics. The SHapley Additive exPlanation method was used to elucidate how each covariate influenced the risk of hypoglycemia. Overall, 419 participants (32.08%) reported at least one hypoglycemic episode. Our findings highlight the non-linear nature of hypoglycemia risk in individuals with T2DM. Insulin therapy, Diabetes duration (> 13.7 years), and eGFR (< 60.2 mL/min/1.73 m2) were the most important predictors of hypoglycemia, followed by age, HbA1C, triglycerides, total cholesterol, gender, and BMI.

A data-driven personalized approach to predict blood glucose levels in type-1 diabetes patients exercising in free-living conditions

OBJECTIVE

The development of new technologies has generated vast amount of data that can be analyzed to better understand and predict the glycemic behavior of people living with type 1 diabetes. This paper aims to assess whether a data-driven approach can accurately and safely predict blood glucose levels in patients with type 1 diabetes exercising in free-living conditions.

METHODS

Multiple machine learning (XGBoost, Random Forest) and deep learning (LSTM, CNN-LSTM, Dual-encoder with Attention layer) regression models were considered. Each deep-learning model was implemented twice: first, as a personalized model trained solely on the target patient's data, and second, as a fine-tuned model of a population-based training model. The datasets used for training and testing the models were derived from the Type 1 Diabetes Exercise Initiative (T1DEXI). A total of 79 patients in T1DEXI met our inclusion criteria. Our models used various features related to continuous glucose monitoring, insulin pumps, carbohydrate intake, exercise (intensity and duration), and physical activity-related information (steps and heart rate). This data was available for four weeks for each of the 79 included patients. Three prediction horizons (10, 20, and 30 min) were tested and analyzed.

RESULTS

For each patient, there always exists either a machine learning or a deep learning model that conveniently predicts BGLs for up to 30 min. The best performing model differs from one patient to another. When considering the best performing model for each patient, the median and the mean Root Mean Squared Error (RMSE) values (across the 79 patients) for predictions made 10 min ahead were 6.99 mg/dL and 7.46 mg/dL, respectively. For predictions made 30 min ahead, the median and mean RMSE values were 16.85 mg/dL and 17.74 mg/dL, respectively. The majority of the predictions output by the best model of each patient fell within the clinically safe zones A and B of the Clarke Error Grid (CEG), with almost no predictions falling into the unsafe zone E. The most challenging patient to predict 30 min ahead achieved an RMSE value of 32.31 mg/dL (with the corresponding best performing model). The best-predicted patient had an RMSE value of 10.48 mg/dL. Predicting blood glucose levels was more difficult during and after exercise, resulting in higher RMSE values on average. Prediction errors during and after physical activity (two hours and four hours after) generally remained within the clinical safe zones of the CEG with less than 0.5% of predictions falling into the harmful zones D and E, regardless of the exercise category.

CONCLUSIONS

Data-driven approaches can accurately predict blood glucose levels in type 1 diabetes patients exercising in free-living conditions. The best-performing model varies across patients. Approaches in which a population-based model is initially trained and then fine-tuned for each individual patient generally achieve the best performance for the majority of patients. Some patients remain challenging to predict with no straightforward explanation of why a patient is more challenging to predict than another.

Chronobiologically-informed features from CGM data provide unique information for XGBoost prediction of longer-term glycemic dysregulation in 8,000 individuals with type-2 diabetes

Type 2 Diabetes causes dysregulation of blood glucose, which leads to long-term, multi-tissue damage. Continuous glucose monitoring devices are commercially available and used to track glucose at high temporal resolution so that individuals can make informed decisions about their metabolic health. Algorithms processing these continuous data have also been developed that can predict glycemic excursion in the near future. These data might also support prediction of glycemic stability over longer time horizons. In this work, we leverage longitudinal Dexcom continuous glucose monitoring data to test the hypothesis that additional information about glycemic stability comes from chronobiologically-informed features. We develop a computationally efficient multi-timescale complexity index, and find that inclusion of time-of-day complexity features increases the performance of an out-of-the-box XGBoost model in predicting the change in glucose across days. These findings support the use of chronobiologically-inspired and explainable features to improve glucose prediction algorithms with relatively long time-horizons.

Artificial Intelligence to Diagnose Complications of Diabetes

Artificial intelligence (AI) is increasingly being used to diagnose complications of diabetes. Artificial intelligence is technology that enables computers and machines to simulate human intelligence and solve complicated problems. In this article, we address current and likely future applications for AI to be applied to diabetes and its complications, including pharmacoadherence to therapy, diagnosis of hypoglycemia, diabetic eye disease, diabetic kidney diseases, diabetic neuropathy, diabetic foot ulcers, and heart failure in diabetes.Artificial intelligence is advantageous because it can handle large and complex datasets from a variety of sources. With each additional type of data incorporated into a clinical picture of a patient, the calculation becomes increasingly complex and specific. Artificial intelligence is the foundation of emerging medical technologies; it will power the future of diagnosing diabetes complications.

Improving Clinical Preparedness: Community Health Nurses and Early Hypoglycemia Prediction in Type 2 Diabetes Using Hybrid Machine Learning Techniques

OBJECTIVES

The aim of the study was to analyze the data of diabetic patients regarding warning signs of hypoglycemia to predict it at an early stage using various novel machine learning (ML) algorithms. Individual interviews with diabetic patients were conducted over 6 months to acquire information regarding their experience with hypoglycemic episodes.

DESIGN

This information included warning signs of hypoglycemia, such as incoherent speech, exhaustion, weakness, and other clinically relevant cases of low blood sugar. Researchers used supervised, unsupervised, and hybrid techniques. In supervised techniques, researchers applied regression, while in hybrid classification ML techniques were used. In a 5-fold cross-validation approach, the prediction performance of seven models was examined using the area under the receiver operating characteristic curve (AUROC). We analyzed the data of 290 diabetic patients with low blood sugar episodes.

RESULTS

Our investigation discovered that gradient boosting and neural networks performed better in regression, with accuracies of 0.416 and 0.417, respectively. In classification models, gradient boosting, AdaBoost, and random forest performed better overall, with AUC scores of 0.821, 0.814, and 0.821, individually. Precision values were 0.779, 0.775, and 0.776 for gradient boosting, AdaBoost, and random forest, respectively.

CONCLUSION

AdaBoost and Gradient Boosting models, in particular, outperformed all others in predicting the probability of clinically severe hypoglycemia. These techniques enable community health nurses to predict hypoglycemia at an early stage and provide the necessary therapies to patients to prevent complications resulting from hypoglycemia.

The Use of a Single Risk Assessment Tool for Mortality and Numerous Hospital-Acquired Conditions

ABSTRACT

Quality assessment organizations leverage numerous patient safety measures to evaluate hospital performance, resulting in significant financial, administrative, and operational burdens on health systems. Low-intensity approaches that allow for reliable risk stratification of patients can reduce the required investment. The Braden score is a routinely performed bedside nursing evaluation validated to assess risk for hospital-acquired pressure injury. We hypothesized that the tool can be used to evaluate risk for other hospital-related adverse outcomes, including mortality, catheter-associated urinary tract infection (CAUTI), and central line-associated bloodstream infection (CLABSI). We found that abnormal Braden scores have significant association with numerous adverse outcome measures, including mortality, CLABSI, CAUTI, and iatrogenic hypoglycemia. Because of its frequency of reevaluation, we have found preliminary evidence that leveraging this tool can reduce harm by quickly identifying the most at-risk patients for various types of iatrogenic harm. We conclude that in the face of increasing automation and technical applications, for example, artificial intelligence-driven tools, highly reliable clinician bedside physical examination and evaluation can still have significant, low-cost, and high-value impact in improving patient safety.

Proactive Identification of Patients with Diabetes at Risk of Uncontrolled Outcomes during a Diabetes Management Program: Conceptualization and Development Study Using Machine Learning

BACKGROUND

The growth in the capabilities of telehealth have made it possible to identify individuals with a higher risk of uncontrolled diabetes and provide them with targeted support and resources to help them manage their condition. Thus, predictive modeling has emerged as a valuable tool for the advancement of diabetes management.

OBJECTIVE

This study aimed to conceptualize and develop a novel machine learning (ML) approach to proactively identify participants enrolled in a remote diabetes monitoring program (RDMP) who were at risk of uncontrolled diabetes at 12 months in the program.

METHODS

Registry data from the Livongo for Diabetes RDMP were used to design separate dynamic predictive ML models to predict participant outcomes at each monthly checkpoint of the participants' program journey (month-n models) from the first day of onboarding (month-0 model) up to the 11th month (month-11 model). A participant's program journey began upon onboarding into the RDMP and monitoring their own blood glucose (BG) levels through the RDMP-provided BG meter. Each participant passed through 12 predicative models through their first year enrolled in the RDMP. Four categories of participant attributes (ie, survey data, BG data, medication fills, and health signals) were used for feature construction. The models were trained using the light gradient boosting machine and underwent hyperparameter tuning. The performance of the models was evaluated using standard metrics, including precision, recall, specificity, the area under the curve, the F1-score, and accuracy.

RESULTS

The ML models exhibited strong performance, accurately identifying observable at-risk participants, with recall ranging from 70% to 94% and precision from 40% to 88% across the 12-month program journey. Unobservable at-risk participants also showed promising performance, with recall ranging from 61% to 82% and precision from 42% to 61%. Overall, model performance improved as participants progressed through their program journey, demonstrating the importance of engagement data in predicting long-term clinical outcomes.

CONCLUSIONS

This study explored the Livongo for Diabetes RDMP participants' temporal and static attributes, identification of diabetes management patterns and characteristics, and their relationship to predict diabetes management outcomes. Proactive targeting ML models accurately identified participants at risk of uncontrolled diabetes with a high level of precision that was generalizable through future years within the RDMP. The ability to identify participants who are at risk at various time points throughout the program journey allows for personalized interventions to improve outcomes. This approach offers significant advancements in the feasibility of large-scale implementation in remote monitoring programs and can help prevent uncontrolled glycemic levels and diabetes-related complications. Future research should include the impact of significant changes that can affect a participant's diabetes management.

Data-based modeling for hypoglycemia prediction: Importance, trends, and implications for clinical practice

Background and objective

Hypoglycemia is a key barrier to achieving optimal glycemic control in people with diabetes, which has been proven to cause a set of deleterious outcomes, such as impaired cognition, increased cardiovascular disease, and mortality. Hypoglycemia prediction has come to play a role in diabetes management as big data analysis and machine learning (ML) approaches have become increasingly prevalent in recent years. As a result, a review is needed to summarize the existing prediction algorithms and models to guide better clinical practice in hypoglycemia prevention.

Materials and methods

PubMed, EMBASE, and the Cochrane Library were searched for relevant studies published between 1 January 2015 and 8 December 2022. Five hypoglycemia prediction aspects were covered: real-time hypoglycemia, mild and severe hypoglycemia, nocturnal hypoglycemia, inpatient hypoglycemia, and other hypoglycemia (postprandial, exercise-related).

Results

From the 5,042 records retrieved, we included 79 studies in our analysis. Two major categories of prediction models are identified by an overview of the chosen studies: simple or logistic regression models based on clinical data and data-based ML models (continuous glucose monitoring data is most commonly used). Models utilizing clinical data have identified a variety of risk factors that can lead to hypoglycemic events. Data-driven models based on various techniques such as neural networks, autoregressive, ensemble learning, supervised learning, and mathematical formulas have also revealed suggestive features in cases of hypoglycemia prediction.

Conclusion

In this study, we looked deep into the currently established hypoglycemia prediction models and identified hypoglycemia risk factors from various perspectives, which may provide readers with a better understanding of future trends in this topic.

Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes

Artificial intelligence can use real-world data to create models capable of making predictions and medical diagnosis for diabetes and its complications. The aim of this commentary article is to provide a general perspective and present recent advances on how artificial intelligence can be applied to improve the prediction and diagnosis of six significant complications of diabetes including (1) gestational diabetes, (2) hypoglycemia in the hospital, (3) diabetic retinopathy, (4) diabetic foot ulcers, (5) diabetic peripheral neuropathy, and (6) diabetic nephropathy.

Hospital Diabetes Meeting 2022

The annual Virtual Hospital Diabetes Meeting was hosted by Diabetes Technology Society on April 1 and April 2, 2022. This meeting brought together experts in diabetes technology to discuss various new developments in the field of managing diabetes in hospitalized patients. Meeting topics included (1) digital health and the hospital, (2) blood glucose targets, (3) software for inpatient diabetes, (4) surgery, (5) transitions, (6) coronavirus disease and diabetes in the hospital, (7) drugs for diabetes, (8) continuous glucose monitoring, (9) quality improvement, (10) diabetes care and educatinon, and (11) uniting people, process, and technology to achieve optimal glycemic management. This meeting covered new technology that will enable better care of people with diabetes if they are hospitalized.

Big Data in Laboratory Medicine—FAIR Quality for AI?

Laboratory medicine is a digital science. Every large hospital produces a wealth of data each day—from simple numerical results from, e.g., sodium measurements to highly complex output of “-omics” analyses, as well as quality control results and metadata. Processing, connecting, storing, and ordering extensive parts of these individual data requires Big Data techniques. Whereas novel technologies such as artificial intelligence and machine learning have exciting application for the augmentation of laboratory medicine, the Big Data concept remains fundamental for any sophisticated data analysis in large databases. To make laboratory medicine data optimally usable for clinical and research purposes, they need to be FAIR: findable, accessible, interoperable, and reusable. This can be achieved, for example, by automated recording, connection of devices, efficient ETL (Extract, Transform, Load) processes, careful data governance, and modern data security solutions. Enriched with clinical data, laboratory medicine data allow a gain in pathophysiological insights, can improve patient care, or can be used to develop reference intervals for diagnostic purposes. Nevertheless, Big Data in laboratory medicine do not come without challenges: the growing number of analyses and data derived from them is a demanding task to be taken care of. Laboratory medicine experts are and will be needed to drive this development, take an active role in the ongoing digitalization, and provide guidance for their clinical colleagues engaging with the laboratory data in research.