To evaluate the correlation between continuous glucose monitoring (CGM)-derived metrics such as time in range (TIR) and time below range (TBR) and clinical parameters in Indian patients with type 2 diabetes mellitus (T2DM), providing insights into the effectiveness of CGM in this population. This is a single-center, retrospective-observational study of T2DM patients aged 18 or older years using the Freestyle Libre/Libre Pro CGM system. Demographic and clinical data of these patients were extracted from medical records, and glycemic metrics from CGM profiles. The study included 248 T2DM patients, of whom 72.98 % were male, with a mean age of 60.48 years and a mean HbA1c value of 8.319 %. Linear regression analysis revealed that TIR was significantly negatively associated with HbA1c values (coefficient=-6.746) and duration of diabetes (coefficient=-0.425). The body mass index (BMI) showed a positive association with TIR (coefficient=0.649; P = 0.032). For TBR, only HbA1c values showed a significant negative association (coefficient=-1.051; P = 0.009), whereas age, duration of diabetes, and BMI were not significant predictors. The study highlights the significant relationships between CGM-derived metrics (TIR and TBR), and clinical parameters in Indian patients with T2DM, and their effectiveness in identifying patients at risk of inadequate glycemic control.
UNSTRUCTURED Adiposology provides a mechanistic framework for phenotyping adipose tissue dysfunction through four pillars and five routine clinical proxies, but it remains cross-sectional and does not specify how the phenotype should be represented, tracked, and acted upon over time within electronic health records (EHRs). We propose the Adiposology Digital Twin (AdDT), a five-layer conceptual architecture that reformulates the published pillars and proxies as time-indexed, longitudinal constructs: baseline phenotyping, longitudinal data acquisition, pillar-level trajectory assessment, trend-aware prioritization, and outcome feedback and refinement. Operating on routinely collected clinical data, without specialized additional data modalities, the AdDT is designed to enable interpretable detection of candidate drift signals at the level of individual mechanistic pillars, trend-aware clinical prioritization, and interoperability with existing EHR infrastructure, without requiring imaging, wearables, or omics. The AdDT is presented as a conceptual architecture and is not a validated clinical tool; retrospective calibration, prospective evaluation, and real-world implementation studies constitute the staged pathway toward clinical use.