Brazilian researchers have developed a model based on artificial intelligence and X-rays that can help quickly identify Covid-19.
According to the official website of the Institute of Electrical and Electronics Engineers (IEEE), With the help of artificial intelligence, X-rays may become an advanced diagnostic tool for identifying people with Covid-19. A group of Brazilian researchers have developed a computer program using a variety of machine learning techniques to be able to detect Covid-19 in recorded X-ray images of the chest with 6.95 to 98.5% accuracy.
Researchers have previously focused on identifying and classifying lung pathologies through medical imaging. Common symptoms of Covid-19, including shortness of breath and cough, can be detected by medical imaging techniques such as CT scans or X-rays.
Victor Hugo C. de Albuquerque, one of the project’s researchers, said: “As the Covid-19 epidemic began, we agreed to use our expertise to help solve this new global problem. Many medical centers do not have enough tests and require a lot of time to process the test; So we focused on improving the tools that are readily available in any hospital and are now being used to diagnose Covid-19. These are X-ray-based tools.
“We decided to investigate whether COVID-19 infection can be automatically detected using X-ray images,” Albuquerque said. X-ray images are readily available and process in less time, unlike nasal swabs or saliva tests.
He added: “Because X-ray imaging is a fast and inexpensive method, it can help patients who are far from large medical centers and do not have access to more sophisticated technologies.” This method can help physicians to automatically identify and classify medical images so that they can measure the severity of the disease and classify its types.
Researchers plan to continue experimenting with a larger database to eventually develop a free online platform for classifying medical images.
The study was published in the IEEE / CAA Journal of Automatica Sinica.