文献速递技术与安全CTMRUS

人工智能驱动的放射学教育:课程规划、实施与评估

Advancing Radiology Education with Artificial Intelligence: Curriculum Planning, Implementation, and Evaluation.
PubMed 文献 · Radiographics · 2026-01-01

本文综述人工智能在放射学教育中的应用潜力,依据Harden十步课程开发框架系统梳理从需求评估到项目管理的各环节。重点介绍生成式模型如何用于教学案例生成、个体化学习路径及结构化评估。作者建议放射学项目从低风险应用起步,逐步建立应对AI整合未来的能力。

要点

  • AI可生成模拟影像病例和放射学试题,扩充教学资源。
  • 依据学员表现数据可构建个体化学习路径,提升教学针对性。
  • 建议从低风险AI应用开始,积累机构实施经验后再扩展。

核心结论

  • AI有望从课程规划、实施到评估各环节提升放射学教育,代表医学教育范式转变。
  • 放射学项目应采取战略且务实的方式引入AI,从低风险应用起步建设机构能力。

研究设计与样本

  • 本文为综述,依据Harden十步课程开发框架系统梳理AI在放射学教育各环节的现有及潜在应用。

临床应用与阅片要点

  • 可应用自然语言处理自动整理学员反馈,辅助需求评估。
  • 教学实践中可使用AI生成教学案例与模拟试题,减少人工备课负担。
  • 采用客观报告比对工具辅助评估学员阅片报告质量,提升评估效率。

局限性

  • 原文指出实施成本高、技术迭代快、存在AI偏倚与错误风险以及数据隐私问题。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

Artificial intelligence (AI) is rapidly integrating into clinical radiology, creating a continuous emphasis on the necessity of teaching its principles to radiology trainees. However, the potential of AI to transform radiology education remains underexplored. The authors review how AI can be leveraged to enhance radiology education, from curriculum planning to its implementation and evaluation. Guided by Harden's 10-step framework for curriculum development, they systematically examine current and potential future applications of AI at each stage. They detail how AI, particularly generative models, can augment traditional educational methods. Such applications include automating needs assessments through natural language processing of learner feedback, personalizing learning pathways based on performance data, and generating diverse educational content, including synthetic imaging cases and radiology board-style questions. Furthermore, AI can enhance teaching strategies through immersive simulations, streamlining assessments with objective report-comparison tools, and improving program management by automating administrative tasks. Although the potential is immense, significant limitations persist, including high implementation costs, the rapid pace of technological change, risks of AI bias and error, and concerns around data privacy. Despite these challenges, AI represents a paradigm shift for medical education. Radiology programs should pursue a strategic and pragmatic approach to AI adoption, starting with low-risk applications to build institutional capacity and prepare the next generation of radiologists for an AI-integrated future. ©RSNA, 2026 Supplemental material is available for this article. See the invited commentary by Tejani and Cook in this issue.

来源:Radiographics,查看原文。版权归原作者所有。

引用

  1. Rouzrokh, Rouzrokh, Khosravi, et al. Advancing Radiology Education with Artificial Intelligence: Curriculum Planning, Implementation, and Evaluation. Radiographics. 2026. PMID: 42531148. doi:10.1148/rg.250198

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。