Nashville, Tenn.-based Vanderbilt University Medical Center used ChatGPT to speed up review and improvement of computer system alerts that are used to support clinical decision-making and found that the technology could provide top-rated suggestions.
Vanderbilt University Medical Center conducted a blind test using seven logic-based alerts currently used at the system to see how ChatGPT's suggestions compared to suggestions from teams of clinical specialists.
A panel of four physicians and one pharmacist reviewed both the suggestions from ChatGPT and from the clinical specialists and found that out of the 20 top-rated suggestions, nine were made from ChatGPT.
"Across healthcare, most of these well-intentioned automated alerts are overridden by busy users. The alerts are seen as serving an essential purpose, but the general need to improve them is clear to everyone," Siru Liu, PhD, a postdoctoral research fellow at VUMC, said in a June 1 press release. "It's apparent to me that AI could help speed this continuing project. ChatGPT appears already highly useful, and with specialized training it could no doubt be made yet more formidable for this vital purpose."