The overall goal of this project is to develop an artificial intelligence tool designed to predict the chief complaints of emergency calls through natural language processing (NLP) and deep learning to automate the identification of an emergency complaint, depicting the outputs in a geospatial information systems (GIS). We plan to develop an artificial intelligence tool designed to predict the chief complaints of emergency calls. We will make use of natural language processing (NLP) and deep learning to automate the identification of an emergency complaint, depicting the outputs in a geospatial information systems (GIS) layer. We will also build an application to deploy this algorithm as a decision-making support tool for emergency calls and for public health planning with data visualization and GIS integrations. We will evaluate the performance of our predictive algorithm through widely known measures of accuracy compared to trained paramedic impressions of chief complaints. Data will be stored on a password-protected and encrypted database on a DUSOM server.

Start Date: 10/12/2020
End Date: 10/12/2022

Main Topic

Emergency, EMS, AI, speech-to-text, emergency classification, GIS,  NLP