A study conducted at the University of Central Florida of America revealed that artificial intelligence systems can diagnose foci of infection with the emerging corona virus in the lung with the same accuracy as human doctors, according to “DPA”.
The study, which was published in the scientific journal Nature Communications, stated that this technology can overcome some of the challenges facing disease diagnostic systems at the present time.
The study team concluded that the algorithmic equation of artificial intelligence can be trained to monitor foci of infection with the Coronavirus in X-ray images with an accuracy of 90 percent, and it can detect positive cases by 84 percent and negative cases by 93 percent.
The “Science Daily” website, a specialist in technology, quoted the researcher, Olas Bagsey, a professor of computer science at the University of Central Florida, as saying that the new algorithm can accurately detect infection with the Corona virus and distinguish it from the regular flu.
In order to conduct the experiment, the artificial intelligence system was trained to discover foci of infection with the Coronavirus in the CT images of 1,280 patients from China, Japan and Italy, and then they were tested on X-ray images of 1337 patients with lung diseases ranging from Corona virus, common pneumonia and lung cancer.
Bagasse said: We have demonstrated that powerful artificial intelligence systems can detect new cases of coronavirus with an accuracy of up to 90 percent.