Development and Validation of a Meal Detection-Based Tool for User Behavior and Meal Management Analysis
Authors' conclusion
This framework enables insights into meal announcement patterns that may support patient counseling, clinical decision support, and large-scale analyses in AHCL systems.
Study context
- Population
- Individuals with type 1 diabetes using hybrid automated insulin delivery (AID) systems; model trained on adults with type 1 diabetes using MiniMed 780G system and evaluated in individuals with type 1 diabetes using advanced hybrid closed-loop (AHCL) systemvalidation cohort
- Participants
- 157 individuals
- Target condition
- Meal events
Performance
As reported in the abstract
Accuracy
96.9%
Precision
≥89.6%
Recall
≥93.9%
Area under the curve
0.996
Further findings
Proportion of on-time announcements and time in range (TIR)
- Correlation of higher proportion of on-time announcements with higher TIR (ρ)
- 0.375
- Increase in TIR per 10% increase in on-time announcements
- 2.66%
- Higher proportion of on-time announcements correlated with lower hyperglycemia
- Q < 0.001
Delay of late announcements and postprandial glycemic outcomes
- Increasing delay of late announcements was associated with worse postprandial glycemic outcomes
Original publication
Development and Validation of a Meal Detection-Based Tool for User Behavior and Meal Management Analysis. J Diabetes Sci Technol. Published 10 November 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →