MedTech Dive July 25, 2024
Researchers at the National Institutes of Health found that a version of ChatGPT analyzed images at an expert level but frequently reached the right answer with incorrect reasoning.
Dive Brief:
- A recent study showed that a version of ChatGPT analyzes medical images at an expert level but frequently reached the right answer with incorrect reasoning.
- The results, which were published Tuesday in the peer-reviewed journal npj Digital Medicine, show OpenAI’s artificial intelligence GPT-4 with vision is as good at answering multiple-choice questions about medical images as physicians who lack access to external resources.
- However, the model made mistakes in image comprehension, while still reaching the right answer, 27% of the time. The researchers said the errors show...