<p>Flooding is a critical global issue, significantly impacting sustainable development and urban health, particularly in rapidly urbanizing regions. In Vietnam, flooding poses severe challenges, with Hanoi being notably affected due to its rapid urban expansion and reduction of green spaces. This study evaluates flood susceptibility in Hanoi using high-resolution remote sensing imagery integrated with machine learning models. The Artificial Neural Network (ANN) model, optimized with a Genetic Algorithm (GA), demonstrated superior performance (R²_test = 0.823; RMSE = 4.332; and MAE = 4.020). The resulting flood susceptibility map highlights stark contrasts between urban and suburban areas. Urban districts such as Cau Giay, Nam Tu Liem, Ha Dong, and Thanh Xuan exhibit high to very high flood risks due to dense construction, high population density, and proximity to rivers. Conversely, suburban areas generally show lower susceptibility, except for densely developed regions like Thach That and Quoc Oai districts. These findings underscore the need for comprehensive flood sensitivity mapping across temporal and spatial dimensions to inform urban planning and management. This research provides a valuable tool for early flood risk detection and supports policymakers in making informed decisions to enhance urban health and resilience.</p>