Beyond the Mean: A Machine Learning-Based Trend Analysis of CGM Metrics for Improved HbA <sub>1C</sub> Estimation in Type 2 Diabetes
BACKGROUND: formation. OBJECTIVE: in type 2 diabetes. METHODS: . From a standardized 90-day CGM window, 51 metrics were extracted. Benchmark models (mean glucose and GMI) were compared with models developed using forward and exhaustive feature selection with threefold cross-validated multiple linear regression. RESULTS: -squared (0.61). Nighttime and hyperglycemia‑related metrics were consistently selected. CONCLUSION: estimation improves when variability and temporal patterns are included. Nighttime hyperglycemia adds notable predictive value, though further validation is needed.
Authors
- Thomas Kronborg (ORCID: https://orcid.org/0000-0002-2693-0527)
- Majken Jensen
- I Hirsch
- Stine Hangaard
- Camilla H. N. Thomsen (ORCID: https://orcid.org/0000-0001-6229-7282)
- David C. Klonoff (ORCID: https://orcid.org/0000-0001-6394-6862)
- Simon L. Cichosz (ORCID: https://orcid.org/0000-0002-3484-7571)
- Peter Vestergaard
Institutions
- Steno Diabetes Centers (DK)
- Novo Nordisk (Denmark) (DK)
- University of Washington (US)
- Aalborg University Hospital (DK)
- Mills Peninsula Health Services (US)
- Aalborg University (DK)
Publication Details
- Journal
- Journal of Diabetes Science and Technology
- Published
- 2026-05-08
- DOI
- https://doi.org/10.1177/19322968261441312
- Primary Topic
- Diabetes Management and Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00