AI-enabled targeting of public health interventions through dynamic characterisation of the environment

The health benefits of diet and physical activity are well known, but the environment plays a crucial role in shaping these behaviours and, in turn, health risks. Previous studies have focused on people’s home addresses, but this approach overlooks where people actually go, when they go, and what they do during their day. As a result, the relationship between environmental factors and personal behaviours is often missed.

To fill this gap, the project will use existing data from over 10,000 adults in England, which includes timestamped records of physical activity, location, and diet. By applying AI techniques, such as autonomous agents working with a knowledge graph, it will combine this data with geospatial information to more accurately measure personal environmental exposures over time.

The findings will be useful for local authorities, providing insights into how people use spaces and the timing and duration of their exposure to health-related environments. In the long term, this research could pave the way for real-time interventions, such as smartphone alerts, to encourage healthier behaviors.