Why the study?
Hypoglycemia in type 1 diabetes can cause symptoms ranging from mild dysphoria to more severe conditions if not detected promptly, requiring innovative detection techniques to identify and prevent episodes.
Do predictive algorithms accurately forecast hypoglycemia in patients with type 1 diabetes?
Population
Patients with type 1 diabetes evaluated across 19 predictive models
Comparison
Algorithmic methodologies for hypoglycemia prediction spanning statistics, machine learning, and deep learning
Design
Systematic review following PRISMA guidelines
Key result
A systematic review of 19 predictive models for type 1 diabetes hypoglycemia found that algorithms using statistics, machine learning, and deep learning achieved accuracies between 70% and 99%.
Authors
Loading...
May enable real-time alerts in T1D care; extends modeling data but leaves open need for outcome trials.
Systematic Review
Do predictive algorithms accurately forecast hypoglycemia in patients with type 1 diabetes?
Predictive models using machine learning and deep learning show satisfactory accuracy (70-99%) for forecasting hypoglycemia in type 1 diabetes, highlighting their potential for integration into mobile health interventions.
Tsichlaki et al. (2021) conducted a systematic review in Type 1 Diabetes. Hypoglycemia prediction algorithms was evaluated on Accuracy of predictive models. A systematic review of 19 predictive models for type 1 diabetes hypoglycemia found that algorithms using statistics, machine learning, and deep learning achieved accuracies between 70% and 99%.