A group of Scottish scientists has discovered a technique to use X-rays to identify the presence of coronavirus disease (Covid-19) infection in a person. The diagnostic test use artificial intelligence (AI) to forecast the existence of the virus within a person.
The test, developed by experts at the University of West Scotland (UWS), is said to be 98% successful. They also claimed that it will be faster than the PCR test, which takes hours to complete.
“There has long been a need for a quick and reliable tool that can detect Covid-19, and this has become even more true with the rise of the Omicron variant,” said Professor Naeem Ramzan, who headed the three-person UWS team.
According to UWS researchers, the new approach uses X-ray equipment to compare scans to a database of roughly 3,000 photos from Covid-19 patients, healthy people, and those with viral pneumonia.
The deep convolutional neural network AI technique then utilizes an algorithm to analyze visual images and produce a diagnosis. According to UWS experts, the approach proved to be more than 98 percent accurate over an exhaustive testing period.
Professor Ramzan stated that many nations are unable to do huge numbers of Covid testing due to insufficient diagnostic instruments. Their research makes use of readily available equipment to detect the infection promptly.
“It could prove crucial, and potentially life-saving, when diagnosing severe cases of the virus, assisting in determining what treatment may be required,” the professor stated.
He did admit, however, that because Covid-19 symptoms are not obvious in X-rays during the early stages of infection, it cannot completely replace PCR testing.
Professor Milan Radosavljevic, Vice-Principal of Research, Innovation, and Engagement at UWS, and another team member stated that the study would now be expanded.