Home > List of Issue > Table of Contents > Abstract

Journal of Nippon Medical School

Full Text of this Article

-Original-

Performance of a Large Language Model on Japanese Emergency Medicine Board Certification Examinations

Yutaka Igarashi1, Kyoichi Nakahara1, Tatsuya Norii2, Nodoka Miyake1, Takashi Tagami3 and Shoji Yokobori1

1Department of Emergency and Critical Care Medicine, Nippon Medical School, Tokyo, Japan
2Department of Emergency Medicine, University of New Mexico, NM, United States of America
3Department of Emergency and Critical Care Medicine, Nippon Medical School Musashi Kosugi Hospital, Kanagawa, Japan


Background: Emergency physicians need a broad range of knowledge and skills to address critical medical, traumatic, and environmental conditions. Artificial intelligence (AI), including large language models (LLMs), has potential applications in healthcare settings; however, the performance of LLMs in emergency medicine remains unclear.
Methods: To evaluate the reliability of information provided by ChatGPT, an LLM was given the questions set by the Japanese Association of Acute Medicine in its board certification examinations over a period of 5 years (2018-2022) and programmed to answer them twice. Statistical analysis was used to assess agreement of the two responses.
Results: The LLM successfully answered 465 of the 475 text-based questions, achieving an overall correct response rate of 62.3%. For questions without images, the rate of correct answers was 65.9%. For questions with images that were not explained to the LLM, the rate of correct answers was only 52.0%. The annual rates of correct answers to questions without images ranged from 56.3% to 78.8%. Accuracy was better for scenario-based questions (69.1%) than for stand-alone questions (62.1%). Agreement between the two responses was substantial (kappa = 0.70). Factual error accounted for 82% of the incorrectly answered questions.
Conclusion: An LLM performed satisfactorily on an emergency medicine board certification examination in Japanese and without images. However, factual errors in the responses highlight the need for physician oversight when using LLMs.

J Nippon Med Sch 2024; 91: 155-161

Keywords
artificial intelligence, emergency medicine, language, medicine, specialty boards

Correspondence to
Yutaka Igarashi, MD, PhD, Department of Emergency and Critical Care Medicine, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan
igarashiy@nms.ac.jp

Received, August 17, 2023
Accepted, October 4, 2023