医学AI背景下“机器学习”课程实践教学体系重构与实践
Reconstruction and practice of practical teaching system for “Machine Learning” courses under the background of medical AI
以《上海市发展医学人工智能工作方案(2025 — 2027年)》为政策指引,围绕智能影像工程专业“机器学习”课程的实践教学展开改革研究。针对当前教学中存在的临床场景关联弱、实验设计系统性不足、临床思维培养缺失等问题,提出以“临床任务贯通、三维素养递进、医教闭环融合”为核心的教学改革框架。通过引入糖尿病与乳腺癌临床数据集,对线性回归、逻辑回归、决策树、朴素贝叶斯及K-Means聚类等实验进行系统性重构,形成贯穿疾病预测、辅助诊断与亚型探索的递进式教学体系,并在各实验环节有机融入临床可解释性、召回率优先、医学伦理反思等核心素养要求。同时,建立“技术能力”与“医学素养”双维评价体系,实现对工程实践能力与临床思维能力的综合评价。教学实践反馈显示,改革有效提升了学生在真实医疗场景下的算法应用与临床问题解决能力,为“人工智能+医疗”跨学科人才培养提供了可借鉴的教学实施路径。
Guided by the “Shanghai Plan for the Development of Medical Artificial Intelligence (2025 — 2027)”, this paper conducts research on the reform of practical teaching for the “Machine Learning” course in the Intelligent Imaging Engineering major. In response to the current teaching issues such as weak connection with clinical scenarios, insufficient systematic experimental design, and lack of clinical thinking cultivation, a teaching reform framework centered on “clinical task integration, three-dimensional competency progression, and medical education closed-loop integration” is proposed. By introducing clinical datasets of diabetes and breast cancer, experiments such as linear regression, logistic regression, decision trees, naive Bayes, and K-Means clustering are systematically restructured to form a progressive teaching system covering disease prediction, auxiliary diagnosis, and subtype exploration. Core competency requirements such as clinical interpretability, recall rate priority, and medical ethics reflection are organically integrated into each experimental stage. Meanwhile, a dual-dimensional evaluation system of “technical ability” and “medical literacy” is established to achieve comprehensive assessment of engineering practice ability and clinical thinking ability. Feedback from teaching practice shows that the reform has effectively enhanced students’ algorithm application and clinical problem-solving abilities in real medical scenarios, providing a referenceable teaching implementation path for the cultivation of interdisciplinary talents in “artificial intelligence + healthcare”.
| [1] |
上海市人民政府. 上海市人民政府办公厅关于印发《上海市发展医学人工智能工作方案(2025—2027年)》的通知(沪府办发〔2024〕21号)[EB/OL]. (2024-11-23)[2024-12-21]. |
| [2] |
尹雪妍, 钱有程. 大模型驱动的机器学习实践课程教学改革应用[J]. 软件导刊,2025, 24(3): 206-210. |
| [3] |
王建花, 朱凡, 茆姝. 机器学习工程实践教学探索[J]. 实验室研究与探索, 2023,42(9): 206-210. |
| [4] |
王琪. 基于CNN与SVM的孤立性肺结节良恶性分类研究[D]. 哈尔滨: 哈尔滨理工大学, 2020. |
| [5] |
籍新宇. 基于CNN模型的CT影像肺结节检测及良恶性诊断[D]. 沈阳: 中国医科大学, 2023. |
| [6] |
白露露, 姚相宇. 应用型本科机器学习课程教学改革探索与实践[J]. 河南财经学刊, 2024, 38(5): 90-93. |
| [7] |
张晶. 模式识别课程建设与教学实践探索[J]. 科教导刊, 2021(31): 122-125. |
| [8] |
于镝, 李睿, 侯明, |
| [9] |
孙晋永, 刘斯韵, 庞程, |
| [10] |
方艳梅, 李冠彬, 卢伟. 机器学习高阶课程建设的探索与实践[J]. 计算机教育, 2023(3): 119-123. |
| [11] |
张亚茹, 郝定溢. 基于项目驱动的“机器学习”课程创新实践探索[J]. 科教导刊, 2024(8): 105-108. |
| [12] |
陈蕾, 张礼. 人工智能背景下机器学习课程教学改革路径探索[J]. 电脑知识与技术, 2024,20(26): 121-124. |
| [13] |
付志鸿, 黄洋. 人工智能专业课程教学的探究与实践: 以“机器学习技术”课程为例[J]. 无线互联科技, 2024, 21(12): 56-59. |
| [14] |
敬超, 郑荣华. 面向研究生的机器学习类课程实践改革研究[J]. 科教导刊, 2024(15): 60-62. |
| [15] |
王捷, 刘云. 医学数据挖掘教学实践: 以基于Weka的机器学习方法为 例[J]. 数字技术与应用, 2024, 42(3): 165-168. |
| [16] |
闫静杰, 朱康, 唐贵进, |
| [17] |
张恒. 面向产学融合的机器学习新型课程建设探索[J]. 科技风, 2020(33): 38-39. |
2026年上海健康医学院教师教学发展研究项目(A1-0200-26-309009-10)
/
| 〈 |
|
〉 |