The National Institutes of Health is turning to artificial intelligence and imaging scans to not only help detect cases of COVID-19 earlier, but to potentially personalize treatments for the spreading disease.
In lung CT scans, the novel coronavirus leaves telltale signs that distinguish it from other respiratory diseases—small white spots and a slightly obscuring haze, described by radiologists as “ground glass,” that indicates fluid build-up and damage to the tissue. Abnormalities are found in the heart scans and ultrasounds of many COVID-19 patients as well.
A collaboration funded by the NIH’s National Institute of Biomedical Imaging and Bioengineering aims to develop new diagnostics and machine learning algorithms to quickly assess the severity of an infection and predict a person’s responses to different treatments.
“This program is particularly exciting because it will give us new ways to rapidly turn scientific findings into practical imaging tools that benefit COVID-19 patients,” NIBIB Director Bruce Tromberg said. “It unites leaders in medical imaging and artificial intelligence from academia, professional societies, industry and government to take on this important challenge.”
The resulting Medical Imaging and Data Resource Center will operate a large, open-source repository that will gather COVID-19 chest images from tens of thousands of patients, “allowing researchers to evaluate both lung and cardiac tissue data, ask critical research questions and develop predictive COVID-19 imaging signatures that can be delivered to healthcare providers,” said Guoying Liu, director of the NIBIB’s MRI program.
The database will be hosted at the University of Chicago, and co-led by a trio of medical imaging societies: the American College of Radiology, the Radiological Society of North America and the American Association of Physicists in Medicine.
In addition, the center will support five infrastructure development projects and oversee 12 research projects, covering about 20 university labs—all with an initial focus on COVID-19, but with plans to expand its imaging data services and AI development to other diseases in the future.
“COVID-19 is our immediate target, but the MIDRC will ultimately enable the medical and scientific communities to mobilize images and data for work against other existing diseases and future healthcare threats,” said the University of Washington’s Paul Kinahan, chair of the AAPM’s research committee.