Treating Long COVID May Get a Lot Easier With This New AI Tool
About 10 to 30 percent of people with COVID do not fully recover for several months, or even years. Instead, they experience a phenomenon known as Long COVID, a group of new or ongoing health problems that can affect much of the body from the brain to even the reproductive organs. Worryingly, there may even be potential damage in organs like the lungs that we cannot see because the diagnostic tools are not powerful enough to gather finer details of the virus attack.
This has made monitoring a patient’s lung health after COVID difficult for clinicians – but that could soon change with the help of artificial intelligence. In one newspaper published on Monday in the magazine Natural machine intelligenceResearchers in Saudi Arabia have developed deep lung parenchymal enhancement (DLPE), an AI tool that can uncover abnormalities buried deep in chest CT scans.
The AI works by analyzing images of the patient’s lung parenchyma — the tissues most commonly impacted by COVID and involved in the processing of oxygen and carbon dioxide. DLPE looks for and enhances the appearance of any abnormalities such as lung lesions that don’t look like they should belong. This tool can pick up lung fibrosis lesions (scarring of the lungs during COVID-19 infection) that cannot be seen by conventional x-rays.
“With DLPE, we have demonstrated for the first time that long-term CT lesions can explain such symptoms,” said Xin Gao, a computational biologist at King Abdullah University of Science and Technology. and is the lead researcher on the paper, said in a press release. “Therefore, scleroderma treatments could be very effective in addressing the long-term respiratory complications of COVID-19.”
The researchers believe their AI will help uncover potential lesions inside the lungs and provide a more definitive diagnosis behind persistent COVID symptoms such as shortness of breath or a persistent cough. It can even be a useful diagnostic tool to detect abnormalities or lung dysfunction not seen in people with other respiratory diseases such as pneumonia, tuberculosis, or lung cancer.