Identification of Activity, Insulin, and Dietary Thresholds Associated With Exercise‐Related Hypoglycaemia and Hyperglycaemia in Children With Type 1 Diabetes: A Free‐Living Study Using Explainable Machine Learning
Authors' conclusion
Phase-specific thresholds across exercise and insulin domains associated with exercise-related dysglycaemia were identified.
Study context
- Population
- Children with T1D (11.9 ± 3.3 years) on injections or open-loop pumps, with CGM, accelerometry, self-reported physical activity (PA), diet and insulin data over seven free-living days
- Participants
- 36 children
- Outcome
- Thresholds of PA characteristics, insulin and carbohydrates associated with hypoglycaemia (< 70 mg/dL) and hyperglycaemia (> 180 mg/dL) during PA, early recovery (2-h post-exercise) and overnight
Methodological detailsfree-living, 7 days
- Free-living
- 7 days
Main finding
As reported in the abstract
Thresholds associated with exercise-related dysglycaemia (Shapley analysis)
Model performance across outcomes and phases: AUC
0.65-0.99
Model performance across outcomes and phases: F1-score
0.62-0.95
Observational data: associations, not effects.
Original publication
Identification of Activity, Insulin, and Dietary Thresholds Associated With Exercise‐Related Hypoglycaemia and Hyperglycaemia in Children With Type 1 Diabetes: A Free‐Living Study Using Explainable Machine Learning. Diabetes Obes Metab. Published 2 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →