Beyond Glucose Monitoring: Multi-Analyte Biosensing and AI-integrated Wearable Platforms for Next-Generation Diabetes Management

Diabetes is a growing global health burden, with projections indicating that up to one in eight adults may be affected by 2050. Effective management increasingly relies on continuous glucose monitoring (CGM), yet most currently deployed systems still track a single analyte, providing limited insight into the complex metabolic perturbations that precede complications. Here, we review emerging multi‑analyte and multi‑modal biosensing platforms that extend CGM beyond glucose to include ketones, lactate, insulin, cortisol, electrolytes, and other biomarkers relevant to diabetes and its comorbidities. We summarize recent advances in plasmonic, electrochemical, and optical sensors and their implementation in non- or minimally invasive formats using skin, sweat, tears, and breath. We then discuss how artificial intelligence and machine learning are being integrated with these biosensors to enable predictive modeling, digital twins, and smart closed‑loop insulin delivery. Emphasis is placed on affordability, scalable manufacturing, and interoperability with smartphones and telemedicine infrastructures, which are essential for equitable deployment in low‑ and middle‑income settings. Rather than providing an exhaustive catalog, this review critically evaluates the strengths, limitations, and translational readiness of current technologies, and highlights design principles for next‑generation intelligent, multi‑analyte biosensing systems that can support proactive and personalized diabetes care.