Machine Learning Models for Inpatient Glucose Prediction

Abstract

Purpose of Review Glucose management in the hospital is difficult due to non-static factors such as antihyperglycemic and steroid doses, renal function, infection, surgical status, and diet. Given these complex and dynamic factors, machine learning approaches can be leveraged for prediction of glucose trends in the hospital to mitigate and prevent suboptimal hypoglycemic and hyperglycemic outcomes. Our aim was to review the clinical evidence for the role of machine learning–based models in predicting hospitalized patients’ glucose trajectory. Recent Findings The published literature on machine learning algorithms has varied in terms of population studied, outcomes of interest, and validation methods. There have been tools developed that utilize data from both continuous glucose monitors and large electronic health records (EHRs). With increasing sample sizes, inclusion of a greater number of predictor variables, and use of more advanced machine learning algorithms, there has been a trend in recent years towards increasing predictive accuracy for glycemic outcomes in the hospital setting. While current models predicting glucose trajectory offer promising results, they have not been tested prospectively in the clinical setting. Summary Accurate machine learning algorithms have been developed and validated for prediction of hypoglycemia and hyperglycemia in the hospital. Further work is needed in implementation/integration of machine learning models into EHR systems, with prospective studies to evaluate effectiveness and safety of such clinical decision support on glycemic and other clinical outcomes.

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

“…It is natural to anticipate a decrease in prediction power as the PH increases due to the limited number of available confounding factors in the data used to train the model. A shorter PH may be more useful for prompt clinician intervention, whereas a larger PH increases the prevalence of an outcome and consequently model performance, but may be less useful as a decision support tool ( 42 ). An increase in PH, on the other hand, improves clinical usability of prediction services by extending the time required to take the necessary action during a critical situation, but at the expense of clinical accuracy.…”

“…In brief, current hypoglycemia prediction models are mostly based on clinical parameters, CGM data, or a combination of both. The predictive accuracy varied with the study population, outcome definition, PH definition, modeling technique and model training and validation approaches ( 42 ). As for clinical data-based models, large sample size and sufficient data processing are frequently required to ensure the accuracy and reliability.…”

“…The PH of such models tends to be broad, ranging from predicting short-term hypoglycemic events like inpatient hypoglycemia to long-term hypoglycemic events like SH events months later. It is important to note that a shorter PH may be more useful for prompt clinician intervention, whereas a larger PH increases the prevalence of an outcome and consequently model performance, but may be less useful as a decision support tool ( 42 ).…”

Prediction of Next Glucose Measurement in Hospitalized Patients by Comparing Various Regression Methods: Retrospective Cohort Study

“…In clinical practice, there is growing interest in developing machine learning algorithms to predict hypoglycemia in the inpatient setting. Although many of the published algorithms use categorical variables [ 38 ], consideration should be given to models that quantitatively predict BG, similar to CGM data. Our findings suggest that smaller prediction horizons are correlated more to the next BG measurement compared with longer periods of data, which suggests that clinicians should consider more recent BG measurements when attempting to predict the next BG measurement.…”

Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes

“…Over the past decade, there has been growing interest in leveraging large EHR data sets to develop prediction models for hypoglycemia using ML algorithms. 24 By including large numbers of predictor variables known to influence glucose homeostasis from very large cohorts of patients, AI technologies can fill an evidence gap by identifying and weighing clinical factors that affect glucose levels in ways that would be difficult for clinicians to recognize from clinical experience alone. Various ML techniques have been used to develop prediction models in both non-ICU and ICU settings, including gradient boosting, [25][26][27][28][29][30] random forest classification, 31 recurrent neural net, 27 and logistic regression.…”