Rapid urbanization has caused an increase in the burden of mental illness across the world. However, traditional study methods are difficult to measure the mental health impact of the increasing size of urban environments. In this project, we proposed a machine learning-based approach to measure the effects of the urban streets on mental health. Specifically, we aim to (1) map the effects of the urban streets on citizen’s mental health at a city scale; (2) find out the essential streetscape elements affecting mental health; (3) propose a research-through-design framework to guide urban landscape planning and design for mental health.