Risk Detection for Infectious Diseases
Assessing the possible threat of infectious diseases is severely complicated by complex urban environments, population dynamics, and multi-modal transportation systems. In addition, evaluating the various levels of risk is only the first step required to make intelligent recommendations for restrictions to be imposed to limit the spread of the disease while maximizing business operations and social activities.
What we did
We built an AI model that assesses each point of interest’s risk for propagating the disease further based on various factors such as people’s movements, interactions, and recently detected infection cases. That model is the first component in the software, followed by the recommender component that analyzes the risk in depth to make the right recommendation, such as enhanced restrictions, closure, randomized testing, or even simply notifying the point of interest.
Our solution enabled various stakeholders to assess the status quo and possible future scenarios with the associated possible risks and provide useful information for businesses to understand and assess their situation and act. In addition, one of the major impacts of the work was enabling fine-tuned micro-interventions (i.e., instead of full city closure, it is possible to assess each POI independently). For example, the macro decision of opening or closing all schools is instead intelligently implemented on a school-by-school basis.