This research uses the Social Vulnerability Index, a measure of urban residential and environmental data, along with socioeconomic factors gathered, cleaned, and aggregated by the Center for Spatial Data Science at the University of Chicago to construct spatial models to understand the concentrations of Childhood Asthma emergency department (ED) visits rates in the Chicago Area. Applying Machine Learning models including Random Forest, Ada-boost, and Neural Network, I predicted the Childhood Asthma ED visits rates at census tracts resolution. The results, which are useful in respiratory and epidemiological studies, can provide more information about the conditions of Childhood Asthma than the previous ZIP Code Tabulation Areas (ZCTA) resolution data.