Related Experiment Video
Updated: Jun 15, 2026

05:51
Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health
Published on: February 21, 2025
454
Design and In Silico Evaluation of an Exercise Decision Support System Using Digital Twin Models.
Gavin Young1,2, Robert Dodier2, Joseph El Youssef3
1School of Medicine, Oregon Health & Science University, Portland, OR, USA.
Journal of Diabetes Science and Technology
|February 23, 2024
Summary
An exercise-aware decision support system (exDSS) improved glucose control for type 1 diabetes (T1D) patients during exercise. The system provided personalized recommendations, significantly increasing time in target glucose range and reducing hypoglycemia.
Area of Science:
- Endocrinology
- Biomedical Engineering
- Exercise Physiology
Background:
- Managing blood glucose levels during physical activity is complex for individuals with type 1 diabetes (T1D).
- Existing decision support tools lack personalization and do not differentiate between aerobic and resistance exercise.
- Safe glucose management (70-180 mg/dL) requires considering multiple exercise-related factors.
Purpose of the Study:
- To develop and evaluate an exercise-aware decision support system (exDSS) for T1D management during exercise.
- To provide personalized glucose control recommendations using digital twin technology.
- To help individuals with T1D maintain safe glucose levels and prevent hypoglycemia during and after physical activity.
Main Methods:
- The exDSS was evaluated using real-world data from a large study involving aerobic and resistance exercise.
- Simulations incorporated participant data including heart rate, insulin, and meal intake.
- Glucose responses were modeled during and after 30-minute exercise sessions for 247 participants (676 aerobic, 631 resistance).
Main Results:
- exDSS significantly improved mean time in glucose range for both aerobic (92.3%) and resistance (87.3%) exercise compared to no intervention and clinical guidelines.
- The system substantially reduced time spent in hypoglycemia for both exercise types (aerobic: 5.1%, resistance: 6.6%) compared to no intervention.
- Glucose outcomes with exDSS were comparable to following clinical guidelines, with significant improvements over no intervention.
Conclusions:
- The exDSS tool demonstrated significant improvements in glucose management during and after exercise for individuals with T1D.
- These findings support the clinical evaluation of exDSS for enhancing exercise safety and efficacy in T1D management.
- Personalized, exercise-aware decision support holds promise for improving glucose control in T1D during physical activity.

