Tencent and Chinese scientists use deep learning to predict fatal COVID-19 cases

Tech firms around the world are in overdrive to help battle the coronavirus pandemic. Research showed that 6.5% of COVID-19 patients can suddenly progress to serious illness, and among them the mortality rate can be as high as 49%. Therefore one of the key tasks for health authorities is to identify and treat patients who may develop severe or fatal syndromes early on.

That’s what has been underway at a research team jointly established by Tencent AI Lab and a group of Chinese public health scientists headed by Zhong Nanshan, the country’s senior medical advisor on COVID-19, since its inception in February.

This week, the team unveiled a deep learning-based model that can predict the risk of coronavirus patients developing critical illness. The details were published in Nature Communications, recounting how the lab devised the model based on a cohort of 1,590 patients from 575 medical centers in China, with further validation from 1,393 patients.

The joint lab made the predictor available online, allowing clinical staff around the world to calculate patients’ probability of developing critical illness within 5, 10 and 30 days using 10 clinical variables. While its immediate focus is on COVID-19, the lab’s long-term mission, in its own words, is to “utilize big data and AI for the screening, prevention and control, and warning of outbreaks, respiratory diseases and thoracic diseases.”

Other Chinese tech giants have been taking up similar projects to contain the deadly virus. Using machine and deep learning, Alibaba built a tool for institutions to forecast the spread of COVID-19, with an alleged 90% accuracy rate. Baidu open-sourced an algorithm for viral structural analysis, claiming the process is 120 times faster than the traditional method.

Tencent AI Lab is the tech behemoth’s effort to stay afloat in a race to develop cutting-edge technology alongside money-making businesses such as video games and social networks. Launched in 2016, the research arm works on computer vision, speech recognition, natural language processing and machine learning through a team of 70 research scientists and 300 engineers.

The lab competes with its local counterparts like Alibaba’s Damo Academy and Baidu Research to hunt down the world’s best AI minds. Often that means hiring a well-established scientist to attract young talent. In recent years, Baidu suffered from the departure of Andrew Ng and Lu Qi, and Tencent AI Lab, too, lost its leading figure, Zhang Tong, last year.