GPT-4 can diagnose and triage patients as well as physicians — and health systems should take advantage of the technology, a study found.
The researchers, including some from Los Angeles-based UCLA Health, found GPT-4 comparable to three board-certified physicians in diagnosing and triaging health conditions — to either emergency, nonemergency or self-care — using 45 clinical vignettes. The generative artificial intelligence, which is used in ChatGPT, was also able to do this without bias, according to the study in JMIR Medical Education.
"The findings from our study should be reassuring for patients, because they indicate that large language models like GPT-4 show promise in providing accurate medical diagnoses without introducing racial and ethnic biases," said study senior author Yusuke Tsugawa, MD, PhD, an associate professor of medicine at David Geffen School of Medicine at UCLA, in a Nov. 8 news release. "However, it is also important for us to continuously monitor the performance and potential biases of these models as they may change over time depending on the information fed to them."
The report noted that health systems can use the findings to "introduce conversational AI to improve patient diagnosis and triage efficiently."