In the study published in The Lancet Digital Health May 11, the researchers showed the machine learning program a large number of X-rays and CT scans that were labeled with the patient’s race. The system was then asked the race of unlabeled sets of pictures, and was able to with an over 90 percent accuracy.
“When my graduate students showed me some of the results that were in this paper, I actually thought it must be a mistake,” said Marzyeh Ghassemi, PhD, a coauthor of the paper and MIT professor.
Dr. Ghassemi is hypothesizing that the algorithm may detect melanin content from medical images in a way that human users haven’t realized yet, but says it will require more research to understand.
“We need to take a pause,” Leo Anthony Celi, MD, a coauthor and professor at Harvard Medical School told the Globe. “We cannot rush bringing the algorithms to hospitals and clinics until we’re sure they’re not making racist decisions or sexist decisions.”
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
