A Taiwanese team from a leading university uses AI to detect potential COVID-19 patients
A Taiwanese team has been recognized for offering a tech solution in the battle against the coronavirus pandemic at a COVID-19-themed hackathon organized by the World Health Organization (WHO).
The #BuildforCOVID19 Global Online Hackathon, taking place last month, was made possible with the support of Facebook, AWS, Microsoft, and other technology companies. The event saw the participation of 18,000 innovators from 175 countries, with 1,560 submissions.
Taiwan’s Medchex was picked as one of the 89 highlighted projects by the panel of judges based on the criteria of "viability, potential, and scalability," according to the event’s website.
The Medchex innovation employs artificial intelligence and machine learning technologies to identify whether someone has been infected with the coronavirus through analysis of chest X-rays. Taking just one second to analyze an X-ray image of the chest, the pneumonia detection system achieved 92 percent accuracy based on 1,400 images.
If the result suggests a high-risk patient, an alert will be automatically sent to the doctor, who will determine whether further tests are needed. This mechanism will lend a helping hand to doctors and diagnosis of the disease.
The team behind the project includes teachers and students of the Department of Computer Science and Information Engineering, National Cheng Kung University (NCKU). A radiologist at the NCKU Hospital was also involved in the invention of the system.