Key result
RMRF algorithm accurately classifies nocturnal hypoglycemia risk linked to HbA1c, bedtime glucose, and insulin.
Why the study?
Nocturnal hypoglycemia is common in diabetes and leads to adverse events; identifying associated factors can improve glucose control and patient care.
Population
127 patients with type 1 diabetes evaluated over 2524 nights
Comparison
Repeated measures random forest (RMRF) algorithm vs standard machine learning models
Authors
Loading...
May aid nocturnal hypoglycemia risk alerts in type 1 diabetes; leaves open prospective validation before clinical use.
The proposed RMRF algorithm improves the identification of factors associated with nocturnal hypoglycemia in patients with type 1 diabetes.
Calhoun et al. (2020) studied type 1 diabetes (n=127). Repeated measures random forest (RMRF) algorithm vs. standard random forest, extremely randomized trees, and generalized linear models was evaluated on nocturnal hypoglycemia. The repeated measures random forest algorithm accurately classified nights at high risk of nocturnal hypoglycemia and identified associations with HbA1c, bedtime blood glucose, and insulin on board.
Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context: