The increase in urban population translates into denser urban traffic and, in turn, more environmental pollution. By taking advantage of the high level of sensorisation in smart cities, pollution levels can be monitored in real time, thus helping citizens to make informed decisions. In this research, we propose a healthy routing system. For this purpose, we estimate the pollution dispersion in the urban area from traffic intensity data. Then, the fastest walking route for a given trip is calculated. This route is evaluated with respect to the pollution that a pedestrian would be exposed to by completing it at a certain time. Finally, the system proposes alternative routes to avoid streets with a higher accumulation of harmful particles. The results demonstrate the feasibility of our proposal for the creation of healthy routes and, furthermore, the usefulness of the system as a real-time pollution estimator.