Analyzing High-Risk Emergency Areas with GIS and Neural Networks: The Case of Athens, Greece

Abstract

Any analysis of health service problems facing the world today must consider that these problems exist in a geographic context. This fact has led to an increased need for accurate and current information to support emergency planning and decision making. In this article we combine geographic information systems (GIS) and neural networks for performing health emergency assessments and generating hazard maps that show areas that are potentially at high risk for emergencies. Through the use of neural networks we predict the location of future emergency events. On these events we use a kernel density estimator to create maps of areas that have a high risk for future emergencies. As a result, emergency services will know in advance where there is a high possibility of an emergency event occurring and can formulate a response, thus improving incident management and health planning. For example, the service can locate ambulances in places near the expected emergency cases, minimizing response time. The approach was tested in stroke-event analysis in the city of Athens, Greece.