39-PUB: A Comprehensive Tool for Analyzing Continuous Glucose Monitoring Data

Introduction & Objective: While existing software packages are available for Continuous Glucose Monitoring (CGM) data analysis, they lack the incorporation of novel CGM-based metrics, specifically focusing on the dynamics of health status, Shannon entropy, and fractal analysis.

Methods: We crafted R functions to execute the computation of essential parameters such as summary statistics, time within/above/below designated glucose thresholds, and glycemic variability indices. Moreover, we introduced novel metrics that capture the dynamics of health status, as well as complexity and fractality analysis. We implemented artificial intelligence methods to provide an integrated approach to diabetes care, incorporating both newly developed and existing CGM-based metrics. To facilitate widespread accessibility, we wrapped these functions into an R package, showcased its utility in analyzing CGM studies involving subjects with Type 1 diabetes and will offer it freely to the public for utilization.

Results: An R package was developed to cover a comprehensive analysis from quality control, exploratory analysis, regular statistical analysis and machine learning modeling on CGM studies. This package has more functionalities than any one of existing CGM analysis tools like GlyCulator, EasyGV (Easy Glycemic Variability), CGM-GUIDE© (Continuous Glucose Monitoring Graphical User Interface for Diabetes Evaluation), GVAP (Glycemic Variability Analyzer Program), Tidepool, CGManalyzer, cgmanalysis, GLU, CGMStatsAnalyser, iglu, rGV, and cgmquantify.

Conclusion: By addressing the limitations of current software programs and introducing advanced features, this software holds the potential to contribute significantly to the field of CGM data analysis and improve the management of diabetes.