Just ahead of the revelation that both President Donald Trump and First Lady Melanie Trump are apparently infected with COVID-19, reports surfaced on the use of artificial intelligence (AI) to detect the latest coronavirus strain. These AI solutions are using algorithms in different ways to determine if a potential case exists when someone undergoes an initial exam, and are said to offer an instant determination. While these results may not help prevent the virus, they can help speed up the testing process so that treatment can begin sooner.
Both studies explored the use of chest X-rays to detect cases of COVID-19. One, conducted by researchers at the University of Minnesota and M Health Fairview, a Minneapolis-based healthcare company, used a total of 118,000 X-rays to build its AI algorithm. 100,000 were of negative COVID-19 patients, while the remaining 18,000 were from patients who had tested positive. From there, by analyzing the results, the algorithm was able to identify patterns that could distinctly be attributed to a positive coronavirus case.
The chest X-ray is standard protocol at many hospitals when someone arrives and is thought to be carrying the virus. With the AI solution, which will now be introduced to 12 M Health Fairview hospitals, the results of the X-ray can be provided in a matter of seconds. The program is being facilitated by Epic, a nationwide medical records management company.
Explains Christopher Tignanelli, MD, assistant professor of surgery at the University of Minnesota Medical School, who is also listed as co-lead for the research, “This may help patients get treated sooner and prevent unintentional exposure to COVID-19 for staff and other patients in the emergency department. This can supplement nasopharyngeal swabs and diagnostic testing, which currently face supply chain issues and slow turnaround times across the country.”
The other study, which was previously published in Nature Communications, involved the University of Central Florida (UCF) and a wide range of test subjects. Specifically, 1,280 patients from China, Japan and Italy participated after their chest X-rays were included in the research. Through the researchers’ AI algorithm, they can reportedly detect COVID-19 in patients who have symptoms, as well as those who don’t. The virus can also be detected after the symptoms go away.
Ulas Bagci, Ph.D., of UCF’s Center for Research in Computer Vision and co-author of the study, “We demonstrated that a deep learning-based AI approach can serve as a standardized and objective tool to assist healthcare systems as well as patients. It can be used as a complementary test tool in very specific limited populations, and it can be used rapidly and at large scale in the unfortunate event of a recurrent outbreak.”
The AI-based testing from the second study reportedly has an accuracy rate of 84% in detecting COVID-19 and a 93% accuracy rate in determining a negative result. As this is just the beginning of the research, fine-tuning the process and the data should be able to lead to even more accurate results and rapid testing.