Data-Driven Carbohydrate Counting Accuracy Monitoring: A Personalized Approach

Accurate carbohydrate counting is crucial for type 1 diabetes mellitus patients on intensive insulin therapy to get on-target blood glucose values. So, it fundamental assess their ability estimate meals’ content and, if needed, recommend training. In this context, we propose a personalized data-driven approach monitor the patients’ of meals. The proposed uses data compute safe range error according characteristics each patient and adjust interval patient's daily routines food habits. Initially, method insulin-to-carbohydrate ratio, sensitivity factor, limits, target error, so could train reach goal. Then, app collected life (i.e., glucose, meals carbohydrates content, bolus) initial needs. Preliminary assessment using FDA-approved University Virginia (UVA)/Padova Type Diabetes Simulator shows potential help being aware needs education how accurate they should be achieve suitable levels. Therefore, tool has great asset healthcare professionals patients, improving learning outcomes leading better glycemic control.