Visual Exploration of Regional Factors of the Health of Urban Residents

Abstract

Urban healthcare operations continuously generate large amounts of electronic medical record data. Combined with urban regional factors, a comprehensive data model is established to conduct a multidimensional visual analysis, which is helpful in extracting and analyzing the correlations between regional factors and residents' health. This paper establishes data models for regional factors such as regional diet, industrial economy, climate, and regional epidemics and introduces an Apriori association algorithm to find disease association relationships, a one-mode projection algorithm for bipartite networks to perform disease clustering, and a collaborative filtering recommendation algorithm to predict patients' potential high-risk diseases. Combined with the general analysis process for the visual exploration of regional health impact factors proposed in this paper, a visual analysis system is designed, and through the collaborative interaction of multiple visual interfaces, the multidimensional visual presentation of the association pattern between regional factors and residents' health is achieved. Taking a sample size of 24,626 electronic medical records in a city as an example, the case study proves that the exploration of regional factors such as catering, climate and regional epidemics can provide an analytical basis and decision support for improving the health level of urban residents.