The global trend towards obesity, physical inactivity and energy-dense diets has led to a rapid rise in the prevalence of diabetes. It is estimated that 415 million people live with diabetes, with a projected increase to 642 million by 2040. Practical and cost-effective interventions at national levels are essential to attenuate the global burden of diabetes. Current evidence have shown that glucose control designed to achieve near-normal glycemia may reduce the risk of chronic complications in patients with type 1 diabetes (T1D) and some patients with type 2 diabetes (T2D). However, a critical knowledge gap remains in understanding what persons with diabetes do or do not develop complications, independently of glucose control, and whether there are populations and regions-specific differences in this risk. We hypothesize that populations from different ethnic origin/regions express diverse specific phenotypes of diabetes and its complications. Understanding the heterogeneity in the risk for diabetes complications could enrich the knowledge of the pathophysiologic pathways, help to identify the high-risk population, as well as develop target intervention strategies for primary and secondary prevention of diabetes. The specific objectives are to understand the heterogeneity in the phenotype of chronic diabetes complications among patients with diabetes in the two nations, including (a) describing differences in phenotypes and patterns of long-term complications between two countries; and (b) unveiling clinical and biological biomarkers and using deep learning systems and precision medicine approaches to identify increased risk and/or protection from complications among Chinese and US populations. By tracking the adverse outcomes, we hope to identify mechanism-based therapies for the treatment of diabetes complications so treatment can be individualized and targeted appropriately.