Flash glucose monitoring data analysed by detrended fluctuation function on beta-cell function and diabetes...

Area of Science:

• Endocrinology and Metabolism
• Biomedical Data Science
• Diabetes Research

Background:

• Beta-cell dysfunction is central to diabetes pathogenesis.
• Accurate diabetes classification is crucial for effective treatment.
• Traditional glucose metrics may not fully capture disease complexity.

Purpose of the Study:

• To develop a data-driven glucose fluctuation metric using detrended fluctuation function (DFF).
• To investigate the correlation between this DFF-based metric and beta-cell function (fasting C-peptide).
• To evaluate the metric's utility in classifying type 1 and type 2 diabetes.

Main Methods:

• Utilized flash glucose monitoring data from 78 type 1 and 59 type 2 diabetes participants.
• Applied detrended fluctuation function (DFF) to extract glucose fluctuation patterns.
• Correlated DFF metric with fasting C-peptide and compared with other glucose variability measures.
• Assessed classification performance using receiver operating characteristic (ROC) analysis in discovery and validation cohorts.

Main Results:

• A significant negative correlation was found between the DFF metric and fasting C-peptide (r = -0.667, P < .001), outperforming other common metrics.
• The DFF metric demonstrated strong classification accuracy for type 1 vs. type 2 diabetes (AUCs in discovery: 0.846-0.868; validation: 0.799-0.862).
• High sensitivity and specificity were achieved in both discovery and validation cohorts, indicating robust performance.

Conclusions:

• The proposed DFF-based glucose fluctuation metric shows significant potential for predicting beta-cell function.
• This metric demonstrates capacity for distinguishing between type 1 and type 2 diabetes.
• Further large-scale, multicenter studies are warranted to validate these findings.