A digital twin-enhanced decision support system improves time-in-range in type 1 diabetes: a randomized clinical...

Area of Science:

• Endocrinology and Metabolism
• Biomedical Engineering
• Diabetes Technology

Background:

• Type 1 Diabetes (T1D) management often relies on multiple daily injections or sensor-augmented pumps.
• Decision-support systems (DSSs) offer a cost-effective approach to improve treatment adherence and glycemic control in T1D.
• Current bolus calculators may lack precision in prandial insulin dosing.

Purpose of the Study:

• To evaluate the safety, feasibility, and clinical utility of STUDIA, a novel DSS with a digital twin-enabled simulation-assisted bolus calculator.
• To compare the efficacy of simulation-assisted bolus calculation versus traditional carbohydrate counting for prandial insulin dosing in T1D.

Main Methods:

• A randomized, open-label, parallel-group study involving 28 participants with T1D.
• Participants used either the STUDIA simulation-assisted calculator or traditional carbohydrate counting for prandial insulin dosing.
• Continuous glucose monitoring was performed using Freestyle Libre sensors.

Main Results:

• The simulation-assisted calculator group experienced a 7% increase in time in the target glucose range (70-180 mg/dL) compared to the control group (p < 0.001).
• Hypoglycemia incidence rate was significantly lower in the STUDIA group (RR 0.31, p = 0.022).
• Model performance showed a mean absolute percentage error of 19.2 ± 6.7% at 60 minutes, with most discrepancies in low-risk zones.

Conclusions:

• The STUDIA system demonstrates safety and feasibility for individuals with T1D.
• Simulation-assisted bolus calculation shows potential clinical utility in improving glycemic control and reducing hypoglycemia.
• This DSS represents a promising advancement in T1D management technology.