From Architecture to Outcomes: Mapping the Landscape of Digital Twins for Personalized Diabetes Care—A Scoping Review

<b>Background/Objectives</b>: Digital twins are emerging as a transformative technology in diabetes management, promising a shift from standardized protocols to highly personalized care. This scoping review aims to systematically map the current landscape of digital twin applications in diabetes, synthesizing evidence on their implementation architectures, analytical models, performance metrics, and clinical integration strategies to identify key trends and critical gaps. <b>Methods</b>: A systematic search was conducted across five electronic databases in accordance with PRISMA-ScR guidelines to identify empirical studies on digital twins for diabetes. Data from the selected articles were extracted to analyze bibliographic characteristics, population data, technological components, performance outcomes, and integration levels. A narrative synthesis was performed to map the evidence. <b>Results</b>: Seventeen studies were included, revealing a rapid increase in publications since 2022, with a notable concentration of research in India. The technological architecture shows a convergence toward machine learning models (e.g., LSTM) powered by data from IoT devices and wearables. Certain interventional studies have reported significant clinical impacts, including HbA1c reductions of up to 1.9% and T2DM remission rates as high as 76.5% in one trial. However, major implementation barriers were identified, including fragmented interoperability standards and low rates of full integration into clinical workflows (35.3%). <b>Conclusions</b>: Digital twins are emerging as powerful tools that show the potential to drive significant clinical outcomes in diabetes care. However, to translate this promise into widespread practice, future efforts must focus on overcoming the critical challenges of standardized interoperability and deep clinical integration. Rigorous, independently validated, long-term trials in diverse populations are essential to confirm these promising findings.