Wireless Mobile Communication and Healthcare. 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings

Research Article

COVID-19 Patient Outcome Prediction Using Selected Features from Emergency Department Data and Feed-Forward Neural Networks

Download
59 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-70569-5_21,
        author={Sophie Peacock and Mattia Cinelli and Frank S. Heldt and Lachlan McLachlan and Marcela P. Vizcaychipi and Alex McCarthy and Nadezda Lipunova and Robert A. Fletcher and Anne Hancock and Robert D\'{y}richen and Fernando Andreotti and Rabia T. Khan},
        title={COVID-19 Patient Outcome Prediction Using Selected Features from Emergency Department Data and Feed-Forward Neural Networks},
        proceedings={Wireless Mobile Communication and Healthcare. 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings},
        proceedings_a={MOBIHEALTH},
        year={2021},
        month={7},
        keywords={Machine learning COVID-19 Electronic health records},
        doi={10.1007/978-3-030-70569-5_21}
    }
    
  • Sophie Peacock
    Mattia Cinelli
    Frank S. Heldt
    Lachlan McLachlan
    Marcela P. Vizcaychipi
    Alex McCarthy
    Nadezda Lipunova
    Robert A. Fletcher
    Anne Hancock
    Robert Dürichen
    Fernando Andreotti
    Rabia T. Khan
    Year: 2021
    COVID-19 Patient Outcome Prediction Using Selected Features from Emergency Department Data and Feed-Forward Neural Networks
    MOBIHEALTH
    Springer
    DOI: 10.1007/978-3-030-70569-5_21
Sophie Peacock1, Mattia Cinelli1, Frank S. Heldt1, Lachlan McLachlan1, Marcela P. Vizcaychipi2, Alex McCarthy2, Nadezda Lipunova1, Robert A. Fletcher1, Anne Hancock1, Robert Dürichen1, Fernando Andreotti1, Rabia T. Khan1
  • 1: Sensyne Health plc
  • 2: Chelsea and Westminster Hospital NHS Foundation Trust

Abstract

The severity of COVID-19 varies dramatically, ranging from asymptomatic infection to severe respiratory failure and death. Currently, few prognostic markers for disease outcomes exist, impairing patient triaging and treatment. Here, we train feed-forward neural networks on electronic health records of 819 confirmed SARS-CoV-2 positive patients admitted to a two-site NHS Trust hospital in London, England. To allow early risk assessment, the models ingest data collected in the emergency department (ED) to predict subsequent admission to intensive care, need for mechanical ventilation and in-hospital mortality. We apply univariate selection and recursive feature elimination to find the minimal subset of clinical variables needed for accurate prediction. Our models achieve AUC-ROC scores of 0.78 to 0.87, outperforming standard clinical risk scores. This accuracy is reached with as few as 13% of clinical variables routinely collected within the ED, which increases the practical applicability of such algorithms. Hence, state-of-the-art neural networks can predict severe COVID-19 accurately and early from a small subset of clinical variables.