GlyTwin: Digital Twin for Glucose Control in Type 1 Diabetes Through Optimal Behavioral Modifications Using Patient-Centric Counterfactuals
Researchers have developed GlyTwin, a novel computational framework designed to improve glucose control for individuals with Type 1 Diabetes. This system enhances digital twin technology by incorporating patient-centric counterfactual explanations to simulate optimal behavioral treatments. GlyTwin recommends adjustments to carbohydrate intake and insulin dosing to reduce hyperglycemic events, personalizing interventions based on user preferences. Evaluations on a new dataset of 50 T1D patients showed GlyTwin achieved 85.8% valid explanations and was 87.3% effective in preventing hyperglycemia compared to historical data. AI
IMPACT Offers a new AI-driven approach to personalized diabetes management, potentially reducing long-term health complications.