AI taught to “predict” deaths of COVID-19: calculations are available online

Tech companies around the world are striving to help fight the coronavirus pandemic. Studies have shown that 6.5% of patients with COVID-19 can suddenly progress to a serious stage of the disease, and the death rate among them can reach 49%. Therefore, one of the key challenges for public health authorities is to identify and treat patients who may develop severe or fatal syndromes in the early stages. A group of Chinese researchers, together with the Tencent AI Lab, decided to help solve this problem. The group has been led by Zhong Nanshan, China’s Senior Medical Consultant on COVID-19, since its inception in February. The research results are published by the journal Nature Communications.

This week, a team of scientists unveiled a deep learning model that can predict the risk of critical illness in coronavirus patients. The lab developed a model based on a cohort of 1,590 patients from 575 medical centers in China, followed by validation in 1,393 patients.

The collaborative lab made the predictor available online, allowing clinical staff around the world to calculate the likelihood of patients developing critical illness within 5, 10, and 30 days using ten clinical variables.

While the project’s main focus is on COVID-19, the lab’s long-term mission is to “use big data and AI for screening, prevention and control, and warning of outbreaks, respiratory and chest diseases.”

The Tencent AI Lab is the tech giant’s attempt to stay afloat in the race to advance technology alongside lucrative businesses like video games and social media. Launched in 2016, the research department focuses on computer vision, speech recognition, natural language processing, and machine learning with a team of 70 scientists and 300 engineers.