Answers are grounded in this paper and cite their sources.
Cardiometabolic conditions, including type 2 diabetes mellitus (T2DM), obesity and hypertension, develop through complex interactions among metabolic, physiological and lifestyle factors. Continuous glucose monitoring (CGM) and wearable devices can capture time-resolved information such as glucose patterns, heart rate, physical activity and sleep. Artificial intelligence (AI) methods have been investigated for glucose forecasting, physiological-state estimation, disease screening, prospective risk stratification and management of established conditions. This narrative review examines seventeen primary studies and one systematic review relevant to early cardiometabolic risk assessment, with particular attention to multimodal data integration, temporal modelling, explainable AI (XAI) and personalised preventive guidance. The reviewed primary studies cover digital-twin-based diabetes management, insulin resistance, glucose prediction, multi-task disease prediction, prediabetes screening, explainable obesity assessment, wearable-based blood-pressure assessment and CGM-based long-term risk stratification. GluFormer demonstrates prospective risk stratification from CGM representations and a dietary-data extension. The remaining gap is narrower: the included evidence does not establish one externally validated framework that jointly assesses multiple cardiometabolic conditions using CGM and broader wearable/clinical modalities while providing validated explanations and evaluated personalised preventive guidance. The review discusses validation, label quality, privacy and ethical challenges. Such systems should be evaluated using participant-level data separation, external validation and uncertainty assessment, and their outputs should be communicated as non-diagnostic risk estimates rather than clinical diagnoses.
No takes yet. Share an insight, caveat, or question.
H et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: