AI Solution Studied by 4,300 People Predicts COVID-19 Severe Risk

COVID-19 Prognosis Prediction Solution
Pilot Application at Sejong Community Treatment Center
Part of the 'Digital New Deal' Project

[Asia Economy Reporter Cha Min-young] A medical solution prototype that predicts the severity risk of COVID-19 patients using big data-based artificial intelligence (AI) technology will be pilot-applied at the Sejong Living Treatment Center.

The Ministry of Science and ICT and the National IT Industry Promotion Agency (NIPA) announced on the 19th that they will advance the COVID-19 prognosis prediction solution prototype.

This solution was prepared as part of the 'Digital New Deal' project, the AI Convergence Project (AI+X), to respond to new respiratory infectious diseases. It supports medical staff in reducing the risk of misjudging severe patients as mild cases and helps patients receive timely and appropriate treatment. It is also expected to enable efficient allocation of medical resources such as hospital beds.

A total of 11 institutions, including major medical institutions such as Samsung Medical Center and Chungnam National University Hospital, and AI specialized companies such as Lunit and Acryl, participated and started developing the solution in July 2020, completing the prototype development last August. The solution was trained on 29,988 data points from 4,258 patients.

Currently, Samsung Medical Center, Ilsan Hospital, and others are conducting user convenience and prospective external performance evaluations of the solution.

The Ministry of Science and ICT and Sejong Chungnam National University Hospital will verify the solution's performance at the Sejong Living Treatment Center until November this year. They plan to expand the target diseases and advance the solution aiming for approval from the Ministry of Food and Drug Safety in 2022.

Song Kyung-hee, Director of AI-based Policy at the Ministry of Science and ICT, said, "We will actively support the advancement of the developed COVID-19 prognosis prediction solution prototype through performance verification at living treatment centers to effectively respond to 'With Corona'."