COVID-19 is putting artificial intelligence to the test in terms of its health care applications.
Researchers around the world have been testing medical AI for new reasons that apply to COVID-19 diagnosis, care, treatment, and research. This is putting artificial intelligence technology to the test in a widespread and unique way that has never been seen before.
This focused use of the technology will help researchers to better understand the tech’s possibilities.
At the UC San Diego health system, Dr. Albert Hsiao and his team have already been working on a medical AI program for the last 18 months. This version of artificial technology is meant to help doctors to spot pneumonia on a chest X-ray. That said, with the pandemic crisis, the team shifted gears to see how their research could help doctors in the US in the battle against the coronavirus.
The researchers rapidly pivoted their program’s deployment, which applies dots of color on X-ray images in areas where lung damage or other indicators of pneumonia may be present. This tech has now been applied to over 6,000 chest X-rays and is proving itself to be helpful in diagnosis, according to Dr. Hsiao, the UCSD director of the augmented imaging and artificial intelligence data analytics laboratory.
This is only one example of many research teams using medical AI programs in battling COVID-19.
A number of researchers across the country have been redirecting their artificial intelligence programs toward various efforts having to do with the COVID-19 crisis. For instance, it has been used to help determine which patients are experiencing the highest complications risks and which ones can be safely moved to lower intensity care. The tech uses machine learning to sift through millions of data factors to spot patterns that clinicians would find difficult or potentially impossible to spot.
That said, only a small number of the artificial intelligence algorithms have undergone rigorous testing against existing standard medical procedures. Therefore, while the tech may seem to be helpful and promising, actually rolling out their use in the current pandemic crisis risks adding confusion for health care providers and placing patients in danger, caution some medical AI experts.