基于PBL的人工智能自主学习临床案例库在内分泌学教学中的构建和应用

Development and Implementation of a PBL-Based AI Clinical Case Repository for Self-Directed Learning in Endocrinology Education

摘要: 探讨构建基于问题导向学习(PBL)的人工智能(AI)自主学习临床案例库的理论框架、实施路径及其在医学教育，特别是内分泌学教学中的应用价值。方法：通过系统收集、筛选和标准化处理内分泌领域的真实临床病例，构建结构化的教学案例库。利用自然语言处理(NLP)和深度学习技术，开发一款能够进行智能问答、辅助诊断推理的AI软件平台，将案例库资源整合其中，形成“案例库–AI引擎–学习者”的互动学习模式。结果：成功构建了一个涵盖垂体、甲状腺、肾上腺等十大内分泌系统疾病的案例库，并初步开发了配套的AI自主学习平台。该平台能够根据学习者输入的症状、体征或问题，智能推送相关案例，引导其遵循PBL思路进行鉴别诊断和决策，实现了不受时空限制的个性化、交互式学习模式。结论：基于PBL的AI自主学习临床案例库将经典的PBL/CBL教学理念与前沿的AI技术相结合，有望显著提升学生的临床思维能力、自主学习兴趣和解决复杂临床问题的能力。

Abstract: Objective: To explore the theoretical framework, implementation pathway, and educational value of constructing a problem-based learning (PBL)-oriented, artificial intelligence (AI)-driven self-directed clinical case repository, with a particular focus on its application in endocrinology education. Methods: Real-world clinical cases in the field of endocrinology were systematically collected, screened, and standardized to establish a structured teaching case repository. Natural language processing (NLP) and deep learning techniques were employed to develop an AI platform capable of intelligent question-answering and diagnostic reasoning. The case repository was integrated into this platform, forming an interactive “case repository-AI engine-learner” model. Results: A comprehensive case repository covering ten categories of endocrine system disorders—including pituitary, thyroid, and adrenal diseases—was successfully constructed. A prototype AI-based self-directed learning platform was also developed. This platform can intelligently recommend relevant cases based on learners’ input of symptoms, signs, or questions, guiding them to follow the PBL approach in differential diagnosis and clinical decision-making. It enables personalized and interactive learning without temporal or spatial constraints. Conclusion: The PBL-based AI self-directed clinical case repository integrates classical PBL/CBL teaching concepts with cutting-edge AI technology. It holds promise for substantially enhancing learners’ clinical reasoning, motivation for self-directed learning, and ability to solve complex clinical problems.