获取途径

作者

Dingyuan Luo, Mengke Liu, Runyuan Yu, Yulian Liu, Wenjun Jiang, Qi Fan, Naifeng Kuang, Qiang Gao, Tao Yin, Zuncheng Zheng

作者单位

com. ⁴ Department of Rehabilitation Medicine Center, Affiliated Tai'an Central Hospital, Qingdao University, No. 29, Longtan Road, Taishan District, Tai'an City, 271000, Shandong, China. zxyyzhengzuncheng@12 Affiliations ¹ Department of Rehabilitation Medicine Center, Affiliated Tai'an Central Hospital, Qingdao University, No. 29, Longtan Road, Taishan District, Tai'an City, 271000, Shandong, China. ² Department of Radiology, Affiliated Shandong Provincial Hospital, Shandong First Medical University, Jinan, 250021, Shandong, China. ³ Department of Rehabilitation Medicine Center, Affiliated Tai'an Central Hospital, Qingdao University, No. 29, Longtan Road, Taishan District, Tai'an City, 271000, Shandong, China. yintaokfk@16 com.

刊名

Scientific reports

年份

2025

卷号

Vol.15 No.1

页码

14119

ISSN

2045-2322

关键词

Artificial intelligence ChatGPT Large Language models Medical education Medical licensing examination.

摘要

This study aims to compare and evaluate the performance of GPT-3.5, GPT-4, and GPT-4o in the 2020 and 2021 Chinese National Medical Licensing Examination , exploring their potential value in medical education and clinical applications. Six hundred original test questions from the 2020 and 2021 NMLE were selected and input into GPT-3.5, GPT-4, and GPT-4o for response. The accuracy of the models across different question types and units was recorded and analyzed. Statistical methods were employed...更多

This study aims to compare and evaluate the performance of GPT-3.5, GPT-4, and GPT-4o in the 2020 and 2021 Chinese National Medical Licensing Examination , exploring their potential value in medical education and clinical applications. Six hundred original test questions from the 2020 and 2021 NMLE were selected and input into GPT-3.5, GPT-4, and GPT-4o for response. The accuracy of the models across different question types and units was recorded and analyzed. Statistical methods were employed to compare the performance differences among the three models. GPT-4o demonstrated significantly higher overall accuracy than GPT-4 and GPT-3.5 . In the 2020 and 2021 exams, GPT-4o achieved accuracy rates of 84.2% and 88.2%, respectively, with the highest accuracy observed in questions related to the digestive system , reaching 94.75%. GPT-4 showed moderate performance, while GPT - 3.5 had the lowest accuracy. Additionally, GPT-4o exhibited a clear advantage in complex question formats, such as case analysis questions and standard matching questions . GPT-4o outperformed its predecessors in the NMLE, demonstrating exceptional comprehension and problem-solving abilities in non-English medical examinations. This study provides important insights into the application and promotion of generative AI in medical education and clinical practice.收起

文献类型

期刊