This project will apply deep learning in the research domains of epidemiology, architecture, and remote sensing on multi-scale data, spanning from the level of the individual household to the metropolitan level to identify risk areas of mosquito-borne diseases in East African cities. The urban malaria vector Anopheles stephensi recently started spreading across East Africa, posing a major health risk to urban populations. While the links between housing conditions and mosquito-borne diseases are increasingly recognized, the relation between attributes of urban environments and vector-borne diseases remains understudied, decreasing the efficacy of measures to address health issues in African cities. This project will address the aforementioned knowledge-gap by investigating the relationship between mosquito densities in households and architectural, ecological, and urban form variables in Dar es Salaam, Tanzania. The study will include household surveys of mosquito densities and housing conditions as well as high-resolution geospatial surveys of urban areas. The study will conduct statistical data analyses of the links between mosquito densities in households and indicators derived from the household and neighborhood surveys. The findings will be utilized in automated analysis of multispectral satellite remote sensing imagery based on deep learning models trained with manually delineated attributes of the urban environments to identify risk areas of mosquito-borne diseases at an unprecedented detail. The predictive performance of the risk assessment model will be evaluated and the results will be used to guide policy priorities and interventions in addressing mosquito-borne diseases. The research will be integrated in teaching programs at the Royal Danish Academy and the University of Copenhagen. The project thus has potential to address important knowledge gaps, enhance urban resilience in Africa, and strengthen Danish biomedical data science capacities.