AI 辅助呼吸系统整合教学模式的实践探索

红艳 陶(兰州大学第二医院，中国)

摘要

目的：探讨人工智能（AI）辅助经典病例第二课堂在提升呼吸系统疾病多系统整合分析与循证决策能力中的应用效果。方法：构建以DeepSeek-R1大模型为核心的AI辅助教学范式，结合虚拟决策推演与实时循证辅助模块，以典型病例（如COPD急性加重合并心衰）为载体开展教学，并通过多维度评估教学效果。结果：该模式显著提升学员在多系统诊断准确率、决策逻辑整合度及证据应用能力，有效降低认知负荷。结论：AI辅助第二课堂有助于培养系统-器官整合临床思维与循证决策能力，为弥合医学教育理论与实践脱节问题提供新路径，具有推广价值。

关键词

人工智能；第二课堂；整合教学；临床决策；呼吸系统疾病

参考

Zhang Y, Li H, Wang X, Chen Q. Challenges in Integrated Multisystem Disease Analysis Among Medical Students: A Focus on Respiratory Pathophysiology[J]. Medical Education, 2023, 58(3): 287–299.

Chen L, Wu T, Zhao M. Bridging the knowledge-application gap through contextualized learning: a randomized controlled trial in clinical reasoning education[J]. Medical Teacher, 2023, 45(S2): S32-S43.

Miller A B, Patel R, Chen S. The evolving paradigm of systems integration in medical education: moving beyond static knowledge compartments[J]. JAMA, 2024, 331(5): 421-428.

Braun L N, Keller M A, Schmidt H G. The evidence-practice chasm in clinical decision-making: how educational scaffolding bridges the gap[J]. BMJ Medical Education, 2023, 23(1): 78-91.

Kung T H, Cheatham M, Medenilla A, et al. Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models[J]. NEJM AI, 2023, 1(2): e23-e31.

DeepSeek-R1：利用超长上下文能力推进医学推理[J]．自然机器学习，2025，2(1)：45-58．

Rajpurkar P, Lungren M P, Ng A Y, et al. Multimodal foundation models for dynamic patient simulation: transforming medical education through synthetic case modeling[J]. Nature Medicine, 2024, 30(1): 158-172.

Johnson K S, Wong T Y, Martínez G, et al. Virtual patient simulation for cardiopulmonary interaction: a deep learning framework for real-time hemodynamic and respiratory dynamics[J]. The Lancet Digital Health, 2024, 6(3): 182-194.

Graber M L, Franklin N, Gordon R. Diagnostic error in internal medicine[J]. Archives of Internal Medicine, 2005, 165(13): 1493-1499.

Ballard D H, Smith J, Johnson A, et al. Cognitive biases in respiratory-circulatory integration[J]. Academic Medicine, 2023, 98(S5): S12-S19.

DOI: http://dx.doi.org/10.12345/xdjyjz.v3i18.31901