Automated Incident Location Identification for EMS from Ambulance Geospatial Data

Published in "5th Joint International Conference on Data Science & Management of Data (CODS-COMAD 2022)"
Sruthikeerthi Nandita R , Goutham Zampani , Gokul S Krishnan , Gitakrishnan Ramadurai , B Ravindran

Emergency Medical Services (EMS) are a crucial part of the healthcare system. For evaluating the quality of EMS and better planning of EMS allocation to emergency requests, it is important to know the incident location for each emergency request. Currently in India, the callers requesting EMS verbally describe the incident location to which ambulances are dispatched. Therefore, incident locations are stored as descriptions and not as geo-coordinates. However, most EMS vehicles are equipped with GPS. There is a need for automated approaches to identify the incident locations from geospatial data of ambulances. This paper presents a system that uses a clustering approach followed by a distance heuristic implementation for effective identification of the incident locations from ambulance trajectories. Overall, for various levels of precision set for manual validation, the approach proves to be effective with an accuracy of 87.14%. The identified incident locations can be used for further analysis such as evaluating quality of EMS and planning and optimization of EMS allocation to emergency requests.