Researchers at Boston-based Beth Israel Deaconess Medical Center found that GPT-4 can select the correct challenging medical diagnosis 39 percent of the time and include the correct diagnosis on its list of potential diagnoses 64 percent of the time.
To test the generative artificial intelligence tool the researchers had it analyze data and laboratory findings of 70 challenging medical cases, according to a June 15 Beth Israel Deaconess news release.
The news comes after GPT failed an exam widely used by urologists in training.
"While chatbots cannot replace the expertise and knowledge of a trained medical professional, generative AI is a promising potential adjunct to human cognition in diagnosis," said first author Zahir Kanjee, MD, a hospitalist at Beth Israel Deaconess and assistant professor of medicine at Harvard Medical School. "It has the potential to help physicians make sense of complex medical data and broaden or refine our diagnostic thinking. We need more research on the optimal uses, benefits and limits of this technology, and a lot of privacy issues need sorting out, but these are exciting findings for the future of diagnosis and patient care."