Sleep and Activity Prediction for Type 2 Diabetes Management Using Continuous Glucose Monitoring

Kimmo Karkkainen, Gregory D Lyng, Brian L. Hill, Kailas Vodrahalli, Jeffrey Hertzberg, Eran Halperin

Keywords: continuous glucose monitoring, CGM, type 2 diabetes, T2D, activity, sleep, prediction

Abstract: Continuous glucose monitors (CGMs) generate frequent glucose measurements, and numerous studies suggest that these devices may improve diabetes management. These devices give people with diabetes visibility into how lifestyle factors, i.e., meals, physical activity, sleep, stress, and medication adherence, impact their glucose levels. While earlier studies have shown that individual's actions can influence their CGM data, it has not been clear whether CGM data can provide information about these actions. This is the first study to show on a large cohort that CGM can provide information about sleep and physical activities (as aggregated from an activity tracker). We first train a neural network model to determine the sequence of daily activities from CGM signals, and then extend the model to use additional data, such as individual demographics and medical claims history. Using data from 6,981 participants in a Type 2 diabetes (T2D) management program, we show that a model combining an individual's CGM, demographics, and claims data is highly predictive of sleep (AUROC 0.947, AUPRC 0.884), and moderately predictive of physical activity or certain indicators of physical activity (AUROCs/AUPRCs up to 0.817/0.401), as inferred by an activity tracker. These results show that CGM may have wider utility in diabetes management than previously known.