Alliance for Pandemic Preparedness
April 20, 2020
Early prediction of mortality risk among severe COVID-19 patients using machine learning
Category: Article Summary
Topic: Modeling and Prediction
- Using available clinical data, Hu et al developed a machine learning model to predict the outcomes of COVID-19 patients early on, in which the individual risk score based on the four selected variables (age, high-sensitivity C-reactive protein level, lymphocyte count and d-dimer level) and the corresponding probability of death can be used to assess the mortality risk of COVID-19 patients.
- The authors provide a web link to their predictive model, and suggest its use by clinicians who can determine appropriate interventions early on to improve patient outcomes.
Hu et al. (Apr 19, 2020). Early prediction of mortality risk among severe COVID-19 patients using machine learning. Pre-print downloaded Apr 20 from https://doi.org/10.1101/2020.04.13.20064329