AI Speeds Up Detection of Pneumonia That Is Closely Associated With COVID-19

(Photo : Photo by Seanbatty from Pixabay)

Artificial Intelligence makes its way to the forefront as science battles with COVID-19. This time around, AI is eyed to analyze radiographic images that could help in the diagnosis of symptoms that are believed to be related to COVID-19.

In a report, researchers from the University of Chicago and Argonne National Laboratory have teamed up to produce an AI approach to probe more than 6,000 chest X-rays to check the disease and differentiate between its stages. This program, however, has already been in the making for 18 months and was immediately employed when the pandemic started.

How Does It Work?

The machine uses a new problem-solving technique in which it will allow clinicians to employ an AI to detect pneumonia on chest X-rays. It mentions in the post that this algorithm will use color-coding in replacing stored images to detect pneumonia. The said images were from five patients who were treated for the coronavirus in both China and the United States.

This is quite helpful as there may be patterns in the X-rays, which may be hard for a radiologist to recognize. Albert Hsiao, M.D., Ph.D., spearheaded the development of this machine learning tool that was trained with 22,000 notations from human radiologists. With the utilization of the AI method in clinical research studies, physicians or radiologists could easily get an estimate of a patient's chances to acquire pneumonia at the time and place of patient care.

"Pneumonia can be subtle, especially if it's not your average bacterial pneumonia, and if we could identify those patients early before you can even detect it with a stethoscope, we might be better positioned to treat those at highest risk for severe disease and death," Hsiao explained.

The radiologist also added that these type of diagnostic is trouble-free as chest X-rays are cheaper, the results are faster, and the equipment itself is portable making it easier to clean. Thanks to this research, medical professionals will now be able to determine who among those admitted will need more intensive care and who can be advised to go home.

ALSO READ: People Are Tired of Staying at Home, Crowds Seen on Beaches Over Memorial Day Weekend

For most patients who have succumbed to coronavirus, the main symptom was pneumonia, which is a condition indicated by shortness of breath due to inflammation and fluid buildup. Severe pneumonia will often need a good amount of stay in the hospital, plus the need to be assisted with breathing ventilators. These devices, however, are in quite sought-after, especially with the overflowing cases of COVID-19.

Meanwhile, other companies have also stepped up their game in solving the dilemma of the coronavirus using AI. One example is the tool developed by Northwestern University, which hastens the process of choosing the projects, which show a great sign of future success in finding the treatment or vaccine for the virus.

DON'T MISS THIS: Potential COVID-19 Drug Shows Promise, But Government Says It Needs More Research