Chinese researchers are now using artificial intelligence (AI) technologies to predict COVID-19 patients' risk of developing critical illnesses, which will facilitate the early triage of these patients.

Jointly conducted by researchers from the Guangzhou Institute of Respiratory Health and Tencent AI Lab, the research was published in the journal Nature Communications.

It is important to identify COVID-19 patients at risk of serious illnesses as soon as possible and carry out early interventions.

Based on the data of 1,590 patients from 575 medical centers, researchers developed a deep learning-based survival model that can predict the risk of COVID-19 patients developing critical illnesses within five, 10, and 30 days, based on their clinical characteristics at admission.

The researchers screened 74 clinical characteristics and identified 10 key risk factors that could be combined to predict the development of critical illnesses, including X-ray imaging abnormalities, age, dyspnea, and a number of comorbidities.

A calculation tool for early triage of critically ill COVID-19 patients based on the model has gone online and is freely available to the public.