Faster than PCR Testing with 98% Accuracy... The Problem Is It Cannot Detect Early Infection
(From left) X-ray images of lungs of a healthy person, a COVID-19 patient, and a patient with viral pneumonia (Photo by MDPI).
[Asia Economy Senior Reporter Jinsoo Lee] A research team in the UK has succeeded in developing a groundbreaking COVID-19 diagnostic method using X-rays.
According to a paper published by MDPI, a Switzerland-based publisher in Basel that releases open-access scientific journals, researchers at the University of the West of Scotland (UWS) delved into an artificial intelligence (AI) program hoping it would assist medical personnel who cannot use the COVID-19 genetic amplification (PCR) testing method.
This AI program, unlike the PCR test which takes several hours to yield results, can determine COVID-19 infection status within minutes.
X-ray technology is employed here. The program compares and interprets a specific individual's chest X-ray images against approximately 3,000 images of COVID-19 patients, healthy individuals, and viral pneumonia patients. The diagnostic accuracy of this AI program was found to be as high as 98%.
Professor Naem Ramzan of the UWS School of Engineering and Computing, who led the study, stated, "Many countries are unable to conduct widespread COVID-19 testing due to a lack of diagnostic tools," adding, "However, with easily accessible X-ray technology, it is possible to determine COVID-19 infection status within minutes."
However, early-stage COVID-19 symptoms cannot be detected by X-rays. Therefore, this technology cannot completely replace the PCR testing method. The research team is reportedly planning to expand their study further.
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.


