Related Experiment Video
Updated: May 9, 2026
Improving IV Insulin Administration in a Community Hospital
Published on: June 11, 2012
Beyond the Mean: A Machine Learning-Based Trend Analysis of CGM Metrics for Improved HbA1C Estimation in Type 2
Camilla H N Thomsen1,2, Simon L Cichosz1, Thomas Kronborg1,2
1Department of Health Science and Technology, Aalborg University, Gistrup, Denmark.
Insights
Estimating Hemoglobin A1C (HbA1C) using continuous glucose monitoring (CGM) is more accurate when including glucose variability and temporal trends. Nighttime hyperglycemia significantly improves HbA1C prediction in type 2 diabetes.
Area of Science:
- Diabetes management
- Glycemic control monitoring
- Medical data analysis
Background:
- Hemoglobin A1C (HbA1C) is the standard for assessing long-term glycemic control in diabetes.
- Continuous glucose monitoring (CGM) provides glucose management indicator (GMI) as an HbA1C estimate, but often differs from lab values, especially in type 2 diabetes.
- GMI's focus on central tendency overlooks glycemic variability and trends crucial for accurate HbA1C formation.
Purpose of the Study:
- To determine if incorporating CGM-derived metrics for variability, excursions, and temporal trends enhances HbA1C estimation in type 2 diabetes.
- To compare machine learning models with benchmark models (mean glucose, GMI) for HbA1C prediction.
- To identify key CGM metrics that improve HbA1C estimation accuracy.
Main Methods:
- Applied a machine learning framework to three-month CGM data from 159 participants with type 2 diabetes.
- Extracted 51 metrics from 90-day CGM data, ensuring ≥70% data coverage and valid end-of-trial HbA1C.
- Utilized forward and exhaustive feature selection with threefold cross-validated multiple linear regression to build predictive models.
Main Results:
- Benchmark models (mean glucose, GMI) achieved an R-squared of 0.53.
- A model incorporating five CGM metrics (nighttime GMI, night-to-overall mean glucose ratio, GRADE, time in tight range, nighttime time above range) improved R-squared to 0.60.
- The best model, substituting nighttime GRADE for nighttime GMI, reached an R-squared of 0.61, with nighttime and hyperglycemia metrics being key predictors.
Conclusions:
- Integrating CGM-derived variability and temporal patterns significantly improves HbA1C estimation accuracy.
- Nighttime hyperglycemia metrics provide substantial predictive value for HbA1C.
- Further validation is required for these enhanced CGM-based HbA1C estimation models.
Background:
Hemoglobin A1C (HbA1C) is the gold standard for assessing long-term glycemic control in people with diabetes. Increasing use of continuous glucose monitoring (CGM) has led to adoption of the glucose management indicator (GMI) as a CGM‑based HbA1C estimate, but GMI often differs from laboratory HbA1C, especially in type 2 diabetes. This discordance may be associated with the fact that GMI, as a measure of central tendency, fails to capture temporal glycemic trends and variability that relate to HbA1C formation.
Objective:
To evaluate whether combining CGM-derived metrics capturing variability, excursions, and temporal trends improves estimation of laboratory-measured HbA1C in type 2 diabetes.
Methods:
A machine learning framework was applied to CGM data from a three-month randomized trial, including 159 participants with type 2 diabetes. Participants had ≥70% CGM data coverage and valid end-of-trial HbA1C. 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:
Benchmark models yielded R-squared = 0.53. A forward selection model including five metrics (GMI at night, night-to-overall mean glucose ratio, glycemic risk assessment diabetes equation, time in tight range [3.0-7.8 mmol/L], time above range [13.9 mmol/L] at night) improved R-squared to 0.60. The best-performing model (substituting GRADE at night for GMI at night) achieved a similar R-squared (0.61). Nighttime and hyperglycemia‑related metrics were consistently selected.
Conclusion:
Continuous glucose monitoring‑based HbA1C estimation improves when variability and temporal patterns are included. Nighttime hyperglycemia adds notable predictive value, though further validation is needed.
Related Concept Videos
01:22Diabetes Mellitus: Type 2 and Gestational
01:52Regression Toward the Mean
01:25Type II Diabetes Mellitus III: Clinical Manifestations and Diagnosis
01:15Diabetes: Management and Pharmacotherapy
Insulin remains the cornerstone of treatment for most patients with type 1 and many...
01:29Hyperglycemia
01:22Diabetes Mellitus: Overview and Type I Subtype
Type 1 diabetes is an autoimmune disease in which the immune system mistakenly attacks and destroys the insulin-producing beta cells in the pancreas. As a result, the body is unable to produce sufficient insulin, and individuals with...