Hypoglycemia Prediction Using Machine Learning Models for Patients With Type 2 Diabetes

Abstract

Hypoglycemia is a significant adverse outcome in patients with type 2 diabetes and has been associated with increased morbidity, mortality, and cost of care.1 In addition, hypoglycemia is a major limiting factor for the optimization of insulin therapy. In patients with frequent self-monitored blood glucose (SMBG) measurements or those who employ continuous glucose monitors, statistical methods may be used to predict hypoglycemia. For example, Rodbard found that hypoglycemia risk can be estimated using mean, standard deviation, coefficient of variation, and the nature of the glucose distribution.2 Kovatchev et al introduced a measure of BG variability called "average daily risk range," which strongly correlated to hypoglycemia. Most patients with type 2 diabetes have only sparse SMBG data, which do not lend themselves to statistical methods. Our goal is to be able to accurately predict an individual's risk for hypoglycemia using sparse data, and by employing mobile health technology to provide the appropriate preventive actions for patients and caregivers.For predictions to be clinically useful, the accuracy of the prediction should have significant confidence. Predictions should provide a forecast for a time window that is sufficient to enable meaningful preventive interventions. Predictions should be enabled with BG data alone; other clinical information, when available, can be used if the accuracy of the prediction increases. And finally, the prediction algorithm should require only approximately 1 to 2 SMBG values per day, which is typical for patients with type 2 diabetes MethodsWe employed machine learning methods for our prediction algorithms (see Figure 1). Machine learning is useful when there is a large amount of example data and when the rules for prediction are unclear. In the case of hypoglycemia, we felt that though physicians were able to intuitively estimate the risk of hypoglycemia, they weren't able to explain 554260D STXXX10.1177/1932296814554260Journal Background: Minimizing the occurrence of hypoglycemia in patients with type 2 diabetes is a challenging task since these patients typically check only 1 to 2 self-monitored blood glucose (SMBG) readings per day.

Early Detection of Heart Failure Using Electronic Health Records

“…Prediction performance improved for temporal and non-temporal models as the observation window size increased with peak performance occurring between 2 and 3 years of patient data [19]. In a different study, significant differences in the accuracy of hypoglycemia prediction using machine learning models for patients with type 2 diabetes were observed as a function of data density (number of self-monitored blood glucose data points) and prediction window size (1 hour to 24 hours) [20]. Model performance increased with denser data and with smaller prediction window size.…”

Machine Learning and Data Mining Methods in Diabetes Research

“…It is worth noting that DM complications are far less common and severe in people with well-controlled blood glucose levels. Many of those complications have been studied through machine learning and data mining applications [78], [79], [80], [81], [82], [83], [84], [85], [87], [88], [89], [90], [92], [94], [95], [96], [97].…”

“…Machine learning methods, such as Random Forest, support vector machines (SVM), k-nearest neighbor, and naïve Bayes, were used by Sudharsan B et al [94] to predict hypoglycemia among patients with T2D, whereas support vector regression was used by Georga et al [95] for the same reason. Moreover, a comparison of already published algorithms was reported by Jensen [96] in the same framework.…”

Towards Personalization of Diabetes Therapy Using Computerized Decision Support and Machine Learning: Some Open Problems and Challenges

“…Sudharsan et al [112] showed that the detection of hypo-and hyperglycemic events for patients with T2DM is achievable with high accuracy, even if only sparse blood glucose values based on self-monitored blood glucose (SMBG) readings once or twice a day are available. They trained the model with data from approximately 10 weeks.…”