Nonprofit healthcare system Virtua Health recently announced that they’ve implemented new artificial intelligence technology in their process for colon cancer screenings.
Announced last week, the new artificial intelligence tool, known as GI Genius, is now available in all five Virtua hospitals in New Jersey. Led by President and CEO Dennis W. Pullin, Virtua’s latest implementation makes it one of the select few healthcare systems that incorporate A.I. in the colon cancer screening process.
Using 13 million polyp images as a basis for GI Genius, the A.I.tool’s database is expected to help identify healthy tissue from cancerous ones in colon cancer screenings. As the doctor performs the test, GI Genius will look over and place a green square around any potentially dangerous areas for the medical expert to review.
With the use of A.I, Virtua professionals point to previous studies as an indication of GI Genius’ success. Studies have found that, for every one percent increase in the percentage of polyp detection, the risk of a colon cancer diagnosis lowers by 3%.
“GI Genius is like a second set of eyes for the physician during a colonoscopy,” said the medical director of Virtua GI and Digestive Health, Gregory Seltzer, MD, per a statement. “This technology combines the skill and training of the gastroenterologist (GI doctor) with the latest innovation in colon cancer detection.”
As technology develops, A.I. is increasingly being used in a variety of fields, including the medical world, in more ways than one.
Previously, the tool has been used to build antibodies in vaccines, find heart conditions through deep video-based learning and help doctors talk to patients.
Last week, biotech startup Insilico Medicine announced that they would start clinical trials in human patients for an A.I.-generated drug that treats idiopathic pulmonary fibrosis, making it the first A.I. drug to reach this stage of approval.
While A.I. may be making waves in the medical world, some experts warn that the new tools may risk perpetuating racial inequities in the healthcare system.
Created using machine learning, the development of these A.I. systems is often based on information that often underrepresents members of the BIPOC community.
Previous studies have shown that clinical trials, in particular, have failed to include a diverse dataset.
According to a 2021 study, of 1,408 records that studies clinical trials for cancers, only 136 studies met the researchers’ inclusion criteria.
The FDA itself has previously been vocal about the need for diversity in clinical trials, urging people to take part in clinical trials.
“Participants in clinical trials should represent the patients that will use the medical products,” said the FDA in a statement. “This is often not the case—people from racial and ethnic minorities and other diverse groups are underrepresented in clinical research. To achieve health equity so all can benefit from clinical trials, we are committed to taking steps to change this.”
