Estimating Breakfast Characteristics Using Continuous Glucose Monitoring and Machine Learning in Adults With or at Risk of Type 2 Diabetes
Authors' conclusion
We designed an ML framework to automatically estimate the timing of meal events from CGM data in individuals without diabetes and in individuals at risk or with T2D. This could provide a more scalable approach for analyzing postprandial glycemia, increasing the feasibility of CGM-based precision nutrition and diabetes risk assessment applications.
Study context
- Population
- Adults at risk of or living with noninsulin-treated T2D, and adults without diabetes, wearing continuous glucose monitoring (CGM)
- Participants
- 15 vs 36
- Comparison
- Machine learning (ML)-estimated vs annotated breakfast PPGRs
- Nondiabetes data setn = 15
- Validation T2D data setn = 36
- Outcome
- Algorithm-estimated vs annotated breakfast PPGRs
Methodological detailsrandom forest ML algorithm, held-out subset and external validation data set
- Random forest ML algorithm
- Held-out subset and external validation data set
Performance
As reported in the abstract
- No statistically significant differences to annotated PPGRs in incremental area under the curve and glucose rise (both data sets)
- P > .05
- Prebreakfast glucose: small difference in nondiabetes data set
- P = .005
- Prebreakfast glucose: no difference in validation T2D data set
- P = .18
Original publication
Estimating Breakfast Characteristics Using Continuous Glucose Monitoring and Machine Learning in Adults With or at Risk of Type 2 Diabetes. J Diabetes Sci Technol. Published 23 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →