Machine learning models for predicting critical illness risk in hospitalized patients with COVID-19 pneumonia

Abstract

A XGBoost-based based clinical model on admission might be used as an effective tool to identify patients at high risk of critical illness.