Comparison of the advantages and disadvantages of artificial intelligence-and teacher-generated explanations for respiratory system medical examination questions

Yao Yinan; Lü Hanbing; Fu Mengjiao; Qu Jingjing; Zhang Pei; Zhou Jianya

doi:10.3760/cma.j.cn116021-20260206-02298

您当前的位置：

首页 >

文章列表页 >

Comparison of the advantages and disadvantages of artificial intelligence-and teacher-generated explanations for respiratory system medical examination questions

Digital Intelligence MedicalEducation | 更新时间：2026-06-17

- Comparison of the advantages and disadvantages of artificial intelligence-and teacher-generated explanations for respiratory system medical examination questions
- Chinese Journal of Medical Education Research Vol. 25, Issue 6, Pages: 674-680(2026)
- 作者机构：
  
  1.浙江大学医学院附属第一医院呼吸与危重症医学科，杭州 310003
  2.浙江大学医学院，杭州 310058
- 作者简介：
  
  Zhou Jianya, Email: zhoujy@zju.edu.cn
- 基金信息：
  
  Medical Education Research Project of The First Affiliated Hospital, Zhejiang University School of Medicine(zyjg202437;zyjg202518)
- DOI：10.3760/cma.j.cn116021-20260206-02298
  CLC： R-05
- Received：06 February 2026，
  
  Published：20 June 2026
- 稿件说明：
移动端阅览
姚一楠, 吕寒冰, 傅梦姣, 等. AI与教师在医学呼吸系统试题解析撰写中的优劣势比较[J]. 中华医学教育探索杂志, 2026, 25(6): 674-680.

Yao Yinan, Lü Hanbing, Fu Mengjiao, et al. Comparison of the advantages and disadvantages of artificial intelligence-and teacher-generated explanations for respiratory system medical examination questions[J]. Chinese Journal of Medical Education Research, 2026, 25(6): 674-680.
姚一楠, 吕寒冰, 傅梦姣, 等. AI与教师在医学呼吸系统试题解析撰写中的优劣势比较[J]. 中华医学教育探索杂志, 2026, 25(6): 674-680. DOI： 10.3760/cma.j.cn116021-20260206-02298.

Yao Yinan, Lü Hanbing, Fu Mengjiao, et al. Comparison of the advantages and disadvantages of artificial intelligence-and teacher-generated explanations for respiratory system medical examination questions[J]. Chinese Journal of Medical Education Research, 2026, 25(6): 674-680. DOI： 10.3760/cma.j.cn116021-20260206-02298.

摘要

目的

本研究旨在对比人工智能与教师在呼吸系统医学试题解析撰写中的优劣势，并分析不同考点类型下的表现差异。

方法

采用问卷调查法，以浙江大学医学院临床医学本科生和浙江大学医学院附属医院的临床呼吸专业带教教师为调查对象。采用相对偏好评分法（5分制），并从专业准确严谨性、考点匹配精准度、结构逻辑清晰度、临床关联实用性及学习引导有效性这5个维度进行评价。数据采用SPSS 26.0软件进行统计分析。计量资料以均数±标准差表示，采用单样本

检验比较总体偏好与中立值（3）的差异，多维度比较采用配对样本

检验，均为双侧检验，检验水准

=0.05。

结果

本研究共回收问卷60份，包含学生问卷38份，教师问卷22份。总体偏好均分为2.94±0.05，低于中立值3（

=-2.30，

=0.025），整体略倾向于AI解析。教师组（2.91±0.09）与学生组（2.95±0.05）均呈轻度倾向于AI，其中学生组与中立值的差异具有统计学意义（

=-2.05，

=0.047）。按不同考点进行亚组分析，诊断组均分为3.08±0.23，略倾向于教师解析；辅助检查组均分为2.87±0.23，略倾向于AI解析；治疗组均分为2.50±0.22，显著倾向于AI解析（

=-2.85，

=0.029）。多维度分析显示总体差异无统计学意义，但在治疗组中，AI解析在专业准确性方面的评分更高。

结论

AI解析在治疗类考点中表现较好，但在依赖临床经验与复杂情境的诊断考点解析上仍不及教师。因此，在实际教学中可根据不同考点特点探索“AI+教师”协同模式，在提升解析效率的同时，加强对临床思维的引导，从而为试题解析的优化提供参考。

Abstract

Objective

To compare the advantages and disadvantages of artificial intelligence (AI) and teachers in generating explanations for respiratory system medical examination questions

and analyze performance differences across knowledge domains.

Methods

A questionnaire-based survey was conducted among undergraduate medical students from Zhejiang University School of Medicine and clinical instructors in respiratory medicine from the affiliated hospitals of Zhejiang University School of Medicine. A relative preference scoring method (5-point scale) was used

and evaluations were performed across five dimensions: professional accuracy and rigor

precision of alignment with key points

structural and logical clarity

clinical relevance and practicality

and effectiveness in learning guidance. Data were analyzed using SPSS 26.0. Continuous variables were expressed as mean ± standard deviation. A one-sample

-test was used to compare overall preference scores with the neutral value of 3. A paired-sample

-test was used for multi-dimensional comparisons. All tests were two-tailed

with a significance level of

=0.05.

Results

A total of 60 questionnaires were collected

including 38 from students and 22 from teachers. The mean overall preference score was 2.94±0.05

which was lower than the neutral value of 3 (

=-2.30

=0.025)

indicating a slight overall preference for AI-generated explanations. Both the teacher group (2.91±0.09) and the student group (2.95±0.05) showed a slight preference for AI

and the score of the student group was significantly different from the neutral value (

=-2.05

=0.047). Subgroup analysis by knowledge domain showed that the diagnosis group scored 3.08±0.23

indicating a slight preference for teacher explanations; the auxiliary examination group scored 2.87±0.23

indicating a slight preference for AI; and the treatment group scored 2.50±0.22

showing a significant preference for AI (

=-2.85

.029). Multi-dimensional analysis revealed no significant overall differences. However

in the treatment group

AI-generated explanations scored higher in professional accuracy.

Conclusions

AI-generated explanations perform better in treatment-related knowledge domains but are inferior to those of teachers in diagnosis-related knowledge domains that rely on clinical experience and involve complex scenarios. Therefore

in teaching practice

an AI + teacher collaborative model can be explored based on the characteristics of different knowledge domains

so as to improve explanation efficiency and strengthen the guidance of clinical thinking

thereby providing a reference for optimizing medical examination question explanations.

关键词

Keywords

references

Ahsan Z . Integrating artificial intelligence into medical education: a narrative systematic review of current applications, challenges, and future directions [J ] . BMC Med Educ , 2025 , 25 ( 1 ): 1187 . DOI: 10.1186/s12909-025-07744-0 http://dx.doi.org/10.1186/s12909-025-07744-0 .

袁靳闲，李金政 . 人工智能应用于医学教育的现状、挑战与前景 [J ] . 中华医学教育探索杂志， 2025 ， 24 （ 6 ）： 777 - 783 . DOI: 10.3760/cma.j.cn116021-20240711-02049 http://dx.doi.org/10.3760/cma.j.cn116021-20240711-02049 .

Cooper A , Rodman A . AI and medical education: a 21st-century Pandora's box [J ] . N Engl J Med , 2023 , 389 ( 5 ): 385 - 387 . DOI: 10.1056/nejmp2304993 http://dx.doi.org/10.1056/nejmp2304993 .

Boscardin CK , Gin B , Golde PB , et al . ChatGPT and generative artificial intelligence for medical education: potential impact and opportunity [J ] . Acad Med , 2024 , 99 ( 1 ): 22 - 27 . DOI: 10.1097/ACM.0000000000005439 http://dx.doi.org/10.1097/ACM.0000000000005439 .

Guo D , Yang D , Zhang H , et al . DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning [J ] . Nature , 2025 , 645 ( 8081 ): 633 - 638 . DOI: 10.1038/s41586-025-09422-z http://dx.doi.org/10.1038/s41586-025-09422-z .

Sami A , Tanveer F , Sajwani K , et al . Medical students' attitudes toward AI in education: perception, effectiveness, and its credibility [J ] . BMC Med Educ , 2025 , 25 ( 1 ): 82 . DOI: 10.1186/s12909-025-06704-y http://dx.doi.org/10.1186/s12909-025-06704-y .

魏嵬，罗高萌，刘润铭，等 . 人工智能辅助教学在医学教育中应用效果的Meta分析 [J ] . 医学新知， 2025 ， 35 （ 10 ）： 1242 - 1248 . DOI: 10.12173/j.issn.1004-5511.202505084 http://dx.doi.org/10.12173/j.issn.1004-5511.202505084 .

Zhou J , Zeng Z , Li L . A meta-analysis of Watson for oncology in clinical application [J ] . Sci Rep , 2021 , 11 : 5792 . DOI: 10.1038/s41598-021-84973-5 http://dx.doi.org/10.1038/s41598-021-84973-5 .

Balaya V , Guimiot F , Bruzzi M , et al . Feasibility of a fetal anatomy 3D atlas by computer-assisted anatomic dissection [J ] . J Gynecol Obstet Hum Reprod , 2020 , 49 ( 9 ): 101880 . DOI: 10.1016/j.jogoh.2020.101880 http://dx.doi.org/10.1016/j.jogoh.2020.101880 .

Holmes W , Bialik M , Fadel C . Artificial intelligence in education: promises and implications for teaching and learning [M ] . Boston : Center for Curriculum Redesign , 2019 .

Zawacki-Richter O , Marín VI , Bond M , et al . Systematic review of research on artificial intelligence applications in higher education:where are the educators? [J ] . Int J Educ Technol High Educ , 2019 , 16 ( 1 ): 39 . DOI: 10.1186/s41239-019-0171-0 http://dx.doi.org/10.1186/s41239-019-0171-0 .

Cukurova M , Suraworachet W , Zhou Q , et al . Towards synergistic teacher-AI interactions with generative artificial intelligence [EB/OL ] . ( 2025-11-24 )[ 2026-02-01 ] . https://arxiv.org/abs/2511.19580 https://arxiv.org/abs/2511.19580 .

杨士凤，孙博实，刘旭云，等 . 基于人工智能辅助诊断平台的结直肠肿瘤规培学员胜任力评价指标体系构建 [J ] . 中华医学教育探索杂志， 2025 ， 24 （ 12 ）： 1592 - 1599 . DOI: 10.3760/cma.j.cn116021-20250122-02148 http://dx.doi.org/10.3760/cma.j.cn116021-20250122-02148 .

Anderson LW , Krathwohl DR , Bloom BS . A taxonomy for learning, teaching, and assessing: a revision of Bloom's taxonomy of educational objectives [M ] . New York : Longman , 2001 .

Shimizu T . DECLARE: a comprehensive, multifaceted cognitive forcing strategy to confront complex cases [J ] . Int J Gen Med , 2023 , 16 : 1505 - 1511 . DOI: 10.2147/ijgm.s406165 http://dx.doi.org/10.2147/ijgm.s406165 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Artificial intelligence-based generation of radiology test items and evaluation of assessment effectiveness through analysis of historical test item bank data

Differences in the application of artificial intelligence tools in domestic and overseas scenarios among medical staff and their implications

Connotations, practices, and challenges of artificial intelligence-empowered medical professionalism education

A survey on the current use and ethical awareness of artificial intelligence among medical trainees

Comparison of core competency frameworks for clinical pharmacists in China and the United States

Related Author

Yao Yinan

Lü Hanbing

Fu Mengjiao

Qu Jingjing

Zhang Pei

Shi Yushu

Qin Xiaoxiao

Liao Jinsheng

Related Institution

Department of Radiology, The First Affiliated Hospital, Zhejiang University School of Medicine

Education Administration Office, The First Affiliated Hospital, Zhejiang University School of Medicine

Department of International Affairs, The First Affiliated Hospital, Zhejiang University School of Medicine

Department of Medical Education, The First Affiliated Hospital, Zhejiang University School of Medicine

Department of Gastroenterology, The First Affiliated Hospital, Zhejiang University School of Medicine

⁰