Chinese General Practice ›› 2025, Vol. 28 ›› Issue (22): 2705-2711.DOI: 10.12114/j.issn.1007-9572.2025.0020

• Special Research Report •     Next Articles

The Construction of Assessment Index System of Artificial Intelligence General Practitioner

  

  1. 1. School of General Medicine and Continuing Education, Capital Medical University, Beijing 100069, China
    2. School of Public Health, Peking University, Beijing 100091, China
    3. Beijing Baichuan Intelligent Technology Co., Ltd, Beijing 100084, China
    4. Vanke School of Public Health and Health, Tsinghua University, Beijing 100084, China
  • Received:2025-02-06 Revised:2025-03-13 Published:2025-08-05 Online:2025-06-30
  • Contact: LIU Min, LIANG Wannian
  • About author:

    ZHAO Yali and LU Xiaoqin are co-first authors

智能全科医生评估指标体系构建

  

  1. 1.100069 北京市,首都医科大学全科医学与继续教育学院
    2.100191 北京市,北京大学公共卫生学院
    3.100084 北京市,北京百川智能科技有限公司
    4.100084 北京市,清华大学万科公共卫生与健康学院
  • 通讯作者: 刘民, 梁万年
  • 作者简介:

    赵亚利与路孝琴为共同第一作者

    作者贡献:

    梁万年、刘民负责课题构思、设计、实施、文章撰写等全程审核和把关;路孝琴、赵亚利、刘珏参与课题实施、文章撰写及修改;张艺帆、朱祖懿、陈开元参与课题实施。

  • 基金资助:
    科技创新2030——"新一代人工智能"重大项目(2021ZD0114100); 清华大学文科建设"双高"计划项目—AI赋能基层医疗服务研究(2024TSG06402)

Abstract:

Background

Developing and promoting intelligent medical assistance information systems is an important means to strengthen the capabilities of general practitioners. It is particularly urgent to standardize and evaluate the performance and effectiveness of intelligent assistance system development.

Objective

To construct an evaluation index system for intelligent general practitioners, and provide scientific tools for evaluating the service ability of intelligent general practitioners.

Methods

From December in 2024 to January in 2025, the evaluation index system for intelligent general practitioners with indicator weights was constructed through literature review and Delphi expert consultation.

Results

The positive coefficient of the three rounds of expert consultation was 100%, the authority coefficient was > 0.8. The coefficients of concordance for the importance of indicators at various levels were 0.210, 0.255, and 0.145, respectively, while those for feasibility were 0.353, 0.245, and 0.150. The final evaluation index system comprised five first-level indicators, 11 second-level indicators, and 47 third-level indicators. The first-level indicators were professional knowledge, basic medical service, active clinical prevention service, physician-patient communication and medical ethics, and education, learning, and research, with weights of 0.169, 0.306, 0.239, 0.145, and 0.141, respectively.

Conclusion

This study established the evaluation indicators for intelligent general practitioners, providing a reference for further standardizing the development and application of intelligent general practitioners. It has significant practical implications for enhancing the clinical practice level of general practitioners through technological empowerment.

Key words: Artificial intelligence, General practitioner, Evaluation index system, Expert consultation

摘要:

背景

开发并推广智能医疗辅助信息系统是加强全科医生服务能力的重要手段,规范并评价智能辅助系统开发的性能和效果尤为迫切。

目的

构建智能全科医生评估指标体系,为评估智能全科医生的服务能力提供科学工具。

方法

2024年12月—2025年1月,通过文献研究和德尔菲法,构建智能全科医生评估指标体系,并确定其指标权重。

结果

三轮专家咨询的积极系数均为100%,权威系数>0.8,各级指标的重要性协调系数分别为0.210、0.255、0.145,可行性协调系数分别为0.353、0.245、0.150,最终确定的考核指标体系包括5个一级指标、11个二级指标、47个三级指标。一级指标分别为专业知识、基本医疗服务能力、预防服务能力、医患沟通与医学伦理、教育学习与科研能力,其权重分别为0.169、0.306、0.239、0.145、0.141。

结论

本研究构建了智能全科医生评估指标,对进一步规范智能全科医生的开发及应用提供参考范围,对科技赋能提高全科医生临床实践水平具有重要现实意义。

关键词: 人工智能, 全科医生, 评估指标体系, 专家咨询

CLC Number: