Politico June 26, 2024
THE LAB
Chat GPT scores a passing grade of 70 percent for analyzing eye images and answering eye health questions correctly, according to a study in JAMA Ophthalmology led by researchers at the University of Toronto.
The chatbot, created by the San Francisco tech company OpenAI, answered 82 percent of multiple-choice questions correctly and 65 percent of the image analysis questions correctly.
Researchers think the software they tested, Chat GPT-4, a version that can analyze images, could do better if it were trained more on eye health data.
Why it matters: The medical field as a whole is looking at using AI to help diagnose and treat patients.
Dr. Marko M. Popovic, an ophthalmologist at the University of Toronto and...
Today's Sponsors
Today's Sponsor