Evaluating AI proficiency in nuclear cardiology: Large language models take on the board preparation exam
GPT-4o demonstrated superior performance among the four LLMs, achieving scores likely within or just outside the range required to pass a test akin to the CBNC examination. Although improvements in medical image interpretation are needed, GPT-4o shows potential to support physicians in answering text-based clinical questions.
Authors
- Kathleen Flood
- Aakash Shanbhag (ORCID: https://orcid.org/0000-0001-5199-3071)
- Panithaya Chareonthaitawee (ORCID: https://orcid.org/0000-0002-3811-8712)
- Robert J.H. Miller (ORCID: https://orcid.org/0000-0003-4676-2433)
- Jamieson M. Bourque (ORCID: https://orcid.org/0000-0003-0727-199X)
- Damini Dey (ORCID: https://orcid.org/0000-0003-2236-6970)
- Lawrence M. Phillips
- Piotr J. Slomka (ORCID: https://orcid.org/0000-0002-6110-938X)
- Valerie Builoff
- Joanna X. Liang
Institutions
- Cedars-Sinai Medical Center (US)
- University of Virginia Health System (US)
- American Society for Radiation Oncology (US)
- Mayo Clinic in Arizona (US)
- Artificial Intelligence in Medicine (Canada) (CA)
- New York University (US)
Publication Details
- Journal
- Journal of Nuclear Cardiology
- Published
- 2024-11-29
- DOI
- https://doi.org/10.1016/j.nuclcard.2024.102089
- Citations
- 14
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 1.68
Funders
- National Heart, Lung, and Blood Institute