A new artificial intelligence (AI) computer software can now even produce doctor's notes so real that two physicians couldn't tell the difference.
In a new study, researchers at the University of Florida and NVIDIA analyzed patient notes made by the new AI software and some written by real doctors. The doctors correctly identified the source of the notes just 49% of the time.
Nineteen researchers claim that their findings, published in Nature, pave the way for AI to help healthcare professionals with previously unheard-of efficiency.
The researchers trained supercomputers to produce medical records using a novel model called GatorTronGPT, which works similarly to ChatGPT.
Hugging Face, an open-source AI website, has received over 430,000 downloads of the GatorTronTM models in their free editions. The study's lead author, Yonghui Wu, claims that GatorTronTM models are the only models on the site that may be used for clinical research.
In health care, everyone is talking about these models. GatorTron™ and GatorTronGPT are unique AI models that can power many aspects of medical research and health care. Yet, they require massive data and extensive computing power to build. We are grateful to have this supercomputer, HiPerGator, from NVIDIA to explore the potential of AI in health care.-Wu
AI and Its Impact on Future Healthcare
Wu and his colleagues created a sizable language model for this study, which enables computers to emulate real human discourse.
Standard writing and discussions are a good fit for these models; nevertheless, medical records provide extra challenges due to their high technicality and the requirement to maintain patient privacy.
This GatorTronGPT model is one of the first major products from UF's initiative to incorporate AI across the university. We are so pleased with how the partnership with NVIDIA is already bearing fruit and setting the stage for the future of medicine.-Elizabeth Shenkman, co-author
One of the many potential applications for a medical GPT is to replace the tediousness of paperwork with notes taken by AI and then transcribed.
Programmers spend weeks equipping supercomputers with clinical vocabulary and language use based on billions of words so that an AI tool may write on par with human writing.
Wu claims that the innovation center is working on a commercial software version.