The accelerating global prevalence of Diabetes Mellitus (DM), particularly in demographically diverse, high-risk regions like the East and West Godavari districts of India, necessitates a paradigm shift from reactive diagnosis to proactive, personalized risk stratification and early complication management. Existing predictive models are generally hindered by two critical deficiencies: their reliance on static, non-regional datasets and their failure to adequately classify the clinically vital 'Pre-Diabetes' stage. This project applies Clinical Big Data analytics, IoT-enabled health monitoring devices, and cloud-based machine learning models to predict and analyze the progression of diabetes and its complications. The project outcomes include the development of advanced machine learning and deep learning frameworks that were validated through international research dissemination in IEEE Conferences. Cloud-based computational infrastructure (AWS) was utilized for large-scale data analysis, and IoT integration supported real-time patient monitoring. The project has successfully demonstrated the role of Artificial Intelligence in predictive healthcare systems.