Our project was supported by the Science and Technology Department of Sichuan Province, China. The project lasted for two years.Vascular cognitive dysfunction is a kind of syndrome characterized by cognitive impairment, which is growing at an alarming rate around the world, and has a great impact on the health quality of residents, causing serious social and medical burdens. Effective early comprehensive intervention measures can significantly reduce the risk of cognitive impairment in high-risk groups. However, there are various risk factors affecting the disease, and the degree of cognitive impairment is different. Traditional epidemiological investigation often involves a lot of manpower and material resources, so it is difficult to implement it regularly and continuously on a large scale. VCI early screening is not a routine physical examination item, and the general population’s understanding and attention to vascular cognitive impairment is far from enough. Cerebrovascular-related diseases are high-incidence chronic non-communicable diseases in Sichuan Province, with vascular cognitive impairment as the main manifestation. According to the report of Sichuan Provincial Health and Wellness Committee in 2021, the incidence rate is 1.36/100,000. Cognitive impairment caused by VCI will consume huge resource costs of patients’ family and society. There is sufficient evidence that there are many risk factors that can lead to VCI, and early intervention can effectively delay or alleviate the symptoms of VCI. With the change of population, the development of economy and the change of people’s lifestyle, the proportion of various risk factors is also changing. At present, relevant early screening methods are urgently needed to investigate and analyze high-risk groups. Based on this, the project team combined with previous research, based on multidimensional real-world data, combined with big data technology, deep learning, decision-making and other technologies, effectively analyzed and processed the data, established a screening model and tested the model performance with an independent verification set, established a feasibility scheme for VCI screening, and explored an early screening scheme for VCI with high expansion value
