Key result
Repeated measures random forest and logistic regression models predicted hypoglycemia during exercise in adults with type 1 diabetes with AUCs of 0.833 and 0.825 and balanced accuracy of 77%.
Why the study?
Exercise increases hypoglycemia risk in type 1 diabetes, but predicting when hypoglycemia occurs during exercise remains a major challenge.
Can repeated measures random forest and logistic regression models accurately predict hypoglycemia during exercise in adults with Type 1 Diabetes?
Observational
Can repeated measures random forest and logistic regression models accurately predict hypoglycemia during exercise in adults with Type 1 Diabetes?
Machine learning models using continuous glucose monitoring and baseline characteristics can accurately predict exercise-induced hypoglycemia in adults with Type 1 Diabetes.
Answers are grounded in this paper and cite their sources.
May aid exercise-related hypoglycemia prediction in T1D; leaves open prospective validation before clinical use.
Objective: Exercise is known to increase the risk for hypoglycemia in type 1 diabetes (T1D) but predicting when it may occur remains a major challenge. The objective of this study was to develop a hypoglycemia prediction model based on a large real-world study of exercise in T1D. Research Design and Methods: Structured study-specified exercise (aerobic, interval, and resistance training videos) and free-living exercise sessions from the T1D Exercise Initiative study were used to build a model for predicting hypoglycemia, a continuous glucose monitoring value <70 mg/dL, during exercise. Repeated measures random forest (RMRF) and repeated measures logistic regression (RMLR) models were constructed to predict hypoglycemia using predictors at the start of exercise and baseline characteristics. Models were evaluated with area under the receiver operating characteristic curve (AUC) and balanced accuracy. Results: RMRF and RMLR had similar AUC (0.833 vs. 0.825, respectively) and both models had a balanced accuracy of 77%. The probability of hypoglycemia was higher for exercise sessions with lower pre-exercise glucose levels, negative pre-exercise glucose rates of change, greater percent time <70 mg/dL in the 24 h before exercise, and greater pre-exercise bolus insulin-on-board (IOB). Free-living aerobic exercises, walking/hiking, and physical labor had the highest probability of hypoglycemia, while structured exercises had the lowest probability of hypoglycemia. Conclusions: RMRF and RMLR accurately predict hypoglycemia during exercise and identify factors that increase the risk of hypoglycemia. Lower glucose, decreasing levels of glucose before exercise, and greater pre-exercise IOB largely predict hypoglycemia risk in adults with T1D.
No takes yet. Share an insight, caveat, or question.
Bergford et al. (2023) conducted an observational in Type 1 diabetes (T1D). Repeated measures random forest (RMRF) and repeated measures logistic regression (RMLR) models was evaluated on Hypoglycemia (continuous glucose monitoring value <70 mg/dL) during exercise (AUC 0.833 (RMRF) vs 0.825 (RMLR)). Repeated measures random forest and logistic regression models predicted hypoglycemia during exercise in adults with type 1 diabetes with AUCs of 0.833 and 0.825 and balanced accuracy of 77%.
Synapse has enriched 10 related papers. Consider them for comparative context: