Medical AI Tool from UF, NVIDIA gets Human Thumbs-Up in First Study

A new artificial intelligence (AI) computer program created by researchers at the University of Florida and NVIDIA can generate doctors' notes so well that two physicians couldn't tell the difference, according to an early study from both groups.

In this proof-of-concept study, physicians reviewed patient notes - some written by actual medical doctors while others were created by the new AI program - and the physicians identified the correct author only 49% of the time.

A team of 19 researchers from NVIDIA and the University of Florida said their findings, published Nov. 16 in the Nature journal npj Digital Medicine, open the door for AI to support health care workers with groundbreaking efficiencies.

The researchers trained supercomputers to generate medical records based on a new model, GatorTronGPT, that functions similarly to ChatGPT. The free versions of GatorTron™ models have more than 430,000 downloads from Hugging Face, an open-source AI website. GatorTron™ models are the site’s only models available for clinical research, according to the article’s lead author Yonghui Wu, Ph.D., from the UF College of Medicine’s department of health outcomes and biomedical informatics.

"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 said.

UF alumnus and NVIDIA co-founder Chris Malachowsky is the namesake of UF’s new Malachowsky Hall for Data Science & Information Technology. A public-private partnership between UF and NVIDIA helped to fund this $150 million structure. In 2021, UF upgraded its HiPerGator supercomputer to elite status with a multimillion-dollar infrastructure package from NVIDIA, the first at a university.

For this research, Wu and his colleagues developed a large language model that allows computers to mimic natural human language. These models work well with standard writing or conversations, but medical records bring additional hurdles, such as needing to protect patients’ privacy and being highly technical. Digital medical records cannot be Googled or shared on Wikipedia.

To overcome these obstacles, the researchers stripped UF Health medical records of identifying information from 2 million patients while keeping 82 billion useful medical words. Combining this set with another dataset of 195 billion words, they trained the GatorTronGPT model to analyze the medical data with GPT-3 architecture, or Generative Pre-trained Transformer, a form of neural network architecture. That allowed GatorTronGPT to write clinical text similar to medical doctors’ notes.

“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,” said Elizabeth Shenkman, Ph.D., a co-author and chair of UF’s department of health outcomes and biomedical informatics.

Of the many possible uses for a medical GPT, one idea involves replacing the tedium of documentation with notes recorded and transcribed by AI. Wu says that UF has an innovation center that is pursuing a commercial version of the software.

For an AI tool to reach such parity with human writing, programmers spend weeks programming supercomputers with clinical vocabulary and language usage based on billions upon billions of words. One resource providing the necessary clinical data is the OneFlorida+ Clinical Research Network, coordinated at UF and representing many health care systems.

“It’s critical to have such massive amounts of UF Health clinical data not only available but ready for AI. Only a supercomputer could handle such a big dataset of 277 billion words. We are excited to implement GatorTron™ and GatorTronGPT models to real-world health care at UF Health,” said Jiang Bian, Ph.D., a co-author and UF Health’s chief data scientist and chief research information officer.

A cross-section of 14 UF and UF Health faculty contributed to this study, including researchers from Research Computing, Integrated Data Repository Research Services within the Clinical and Translational Science Institute, and from departments and divisions within the College of Medicine, including neurosurgery, endocrinology, diabetes and metabolism, cardiovascular medicine, and health outcomes and biomedical informatics.

The study was partially funded by grants from the Patient-Centered Outcomes Research Institute, the National Cancer Institute and the National Institute on Aging.

Peng C, Yang X, Chen A, Smith KE, PourNejatian N, Costa AB, Martin C, Flores MG, Zhang Y, Magoc T, Lipori G, Mitchell DA, Ospina NS, Ahmed MM, Hogan WR, Shenkman EA, Guo Y, Bian J, Wu Y.
A study of generative large language model for medical research and healthcare.
NPJ Digit Med. 2023 Nov 16;6(1):210. doi: 10.1038/s41746-023-00958-w

Most Popular Now

AI in Personalized Cancer Medicine: New …

The application of AI in precision oncology has so far been largely confined to the development of new drugs and had only limited impact on the personalisation of therapies. New...

AI can Predict Brain Cancer Patients…

Artificial Intelligence (AI) can predict whether adult patients with brain cancer will survive more than eight months after receiving radiotherapy treatment. The use of the AI to successfully predict patient outcomes...

Max Planck Institute for Informatics and…

The Max Planck Institute for Informatics and Google deepen their strategic research partnership. With additional financial support from the U.S. IT company, the "Saarbrücken Research Center for Visual Computing, Interaction...

JMIR Medical Informatics Invites Submiss…

JMIR Publications has announced a new section titled, "AI Language Models in Health Care" in JMIR Medical Informatics. This leading peer-reviewed journal is indexed in PubMed and has a unique...

Paper Calls for Patient-First Regulation…

Ever wonder if the latest and greatest artificial intelligence (AI) tool you read about in the morning paper is going to save your life? A new study published in JAMA...

Could ChatGPT Help or Hurt Scientific Re…

Since its introduction to the public in November 2022, ChatGPT, an artificial intelligence system, has substantially grown in use, creating written stories, graphics, art and more with just a short...

Evaluating the Performance of AI-Based L…

A new study evaluates an artificial intelligence (AI)-based algorithm for autocontouring prior to radiotherapy in head and neck cancer. Manual contouring to pinpoint the area of treatment requires significant time...

Making AI a Partner in Neuroscientific D…

The past year has seen major advances in Large Language Models (LLMs) such as ChatGPT. The ability of these models to interpret and produce human text sources (and other sequence...

Chapman Scientists Code ChatGPT to Desig…

Generative artificial intelligence platforms, from ChatGPT to Midjourney, grabbed headlines in 2023. But GenAI can do more than create collaged images and help write emails - it can also design...

DMEA nova Award: Wanted - Visionary Solu…

9 - 11 April 2024, Berlin, Germany. The DMEA nova Award is being presented at DMEA 2024 for the first time. The award honours a digital health startup for an outstanding...

New Digital Therapy Reduces Anxiety and …

A therapist-guided digital cognitive behavioural therapy reduced distress in 89 per cent of participants living with long-term physical health conditions, a new King's College London study finds. Researchers at the Institute...

Europe's Digital Health Industry Me…

9 - 11 April 2024, Berlin, Germany. In just over two months, from 9 to 11 April 2024, DMEA, Europe's leading event for digitalisation of healthcare, will gather digital health experts...