Welcome to the repository for our deep learning model focused on predicting outcomes for out-of-hospital cardiac arrest cases. This project leverages cutting-edge techniques in deep learning to contribute to the field of emergency medicine and improve patient care.
Out-of-hospital cardiac arrest is a critical medical emergency that requires swift and accurate interventions. This repository hosts a deep learning model designed to predict the likelihood of successful resuscitation and patient survival based on a variety of input features. By harnessing the power of data-driven predictions, we aim to assist medical professionals in making informed decisions during high-stress situations.
This work contributes:
- Evaluation of community level information on the predictability of OHCA survival
- Community socioeconomic information including crime, healthcare, and economic factors from public data were merged with CARES
- Baseline results using CARES data achieved an AUROC of 84% use improved to 88% using community information
@article{harford2022utilizing,
title={Utilizing community level factors to improve prediction of out of hospital cardiac arrest outcome using machine learning},
author={Harford, Sam and Darabi, Houshang and Heinert, Sara and Weber, Joseph and Campbell, Teri and Kotini-Shah, Pavitra and Markul, Eddie and Tataris, Katie and Hoek, Terry Vanden and Del Rios, Marina},
journal={Resuscitation},
volume={178},
pages={78--84},
year={2022},
publisher={Elsevier}
}