孙越, 王黎红. 基于AI智能体的医院物价咨询知识闭环管理路径构建. 2026. biomedRxiv.202611.00037
基于AI智能体的医院物价咨询知识闭环管理路径构建
通讯作者: 孙越, suny2656@enzemed.com
DOI:10.12201/bmr.202611.00037
Construction of an AI agent-driven closed-loop management path for hospital pricing consultation knowledge
Corresponding author: 孙越, suny2656@enzemed.com
-
摘要:目的 构建人工智能(artificial intelligence,AI)智能体驱动的医院物价咨询知识闭环管理路径,并探讨其理论逻辑与可行性。方法 以知识转化理论与双环学习理论为框架,采用理论演绎与单场景个案相结合的设计,构建“数据采集—智能研判—答复确认—知识沉淀”四层闭环路径,并在过渡形态(个人电脑受控环境+云端大模型+前置脱敏)下进行原型观察与失败案例回溯编码。结果 路径实现了隐性经验向显性组织知识的转化闭环;过渡形态下观察到初步成效:月均咨询总量由397个降至259个,物价咨询平均响应时长由41.9分钟缩短至8.1分钟,重复咨询占比由12.59%降至4.63%(P<0.001),失败案例触发的纠错性固化条数逐月为5、2、2条,呈知识库收敛趋势。结论 该路径为医院物价咨询知识转化与组织学习提供了可操作框架,初步证据支持其理论合理性与技术可行性,组织效果与长期安全性尚待验证。
Abstract: Abstract: Objective To construct an AI agent-driven closed-loop management path for hospital pricing consultation knowledge, and to explore its theoretical rationale and feasibility. Methods Guided by the SECI knowledge conversion model and double-loop learning theory, a four-layer closed-loop path of “data collection-intelligent analysis-answer confirmation-knowledge consolidation” was constructed through theoretical deduction and a single-setting case study, and was observed in a transitional form (controlled personal-computer environment, cloud-based large language model and pre-deidentification) at a tertiary grade-A hospital in Zhejiang Province, with retrospective coding of failure cases. Results The path converted tacit experience into explicit organizational knowledge. Preliminary improvements were observed: monthly consultations decreased from 397 to 259, mean response time from 41.9 min to 8.1 min, repeated consultations from 12.59% to 4.63% (P<0.001), and corrective solidification triggered by failure cases was 5, 2, and 2 entries in the three months, respectively, indicating knowledge base convergence. Conclusion The path provides an operable framework for knowledge conversion and organizational learning in hospital pricing consultation; its theoretical rationale and technical feasibility are preliminarily supported, while organizational effectiveness and long-term safety require further validation.
Key words: AI agent; large language model; generative artificial intelligence; medical service pricing; billing inquiry提交时间:2026-11-19
版权声明:作者本人独立拥有该论文的版权,预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。 -
图表
-
孟冬晴, 吕文娟. 生成式人工智能赋能医学图书馆知识服务的路径研究. 2025. doi: 10.12201/bmr.202510.00039
陈桂月, 贾春岩, 陈智伟, 白慧君. 生成式人工智能在医学科研管理中的应用进展研究. 2026. doi: 10.12201/bmr.202610.00048
中国科学技术信息研究所, 北京市海淀区复兴路号. 基于生成式人工智能的医学视频创作路径探究. 2025. doi: 10.12201/bmr.202508.00005
向艾宁堃, 田靖雪, 胡德华, 刘海霞. 生成式人工智能对于老年糖尿病应答效能的比较研究. 2025. doi: 10.12201/bmr.202503.00019
马傲, 葛小玲. 人工智能大模型在医疗健康领域应用的研究. 2024. doi: 10.12201/bmr.202408.00039
周澎洋, 刘莉莉, 陈思萱, 张宝林, 蒲旭虹. 基于人工智能大模型的中医电子病历内涵质控系统构建与实践. 2026. doi: 10.12201/bmr.202605.00088
史森中. 生成式人工智能(GAI)基层医疗应用风险及治理策略分析. 2024. doi: 10.12201/bmr.202408.00053
弓孟春, 李雨杭, 马永慧, 弓凯, 刘超, 欧阳自豪, 戴辉. 生成式医学人工智能(GMAI)的临床伦理治理:三维协同路径与中国实践. 2025. doi: 10.12201/bmr.202511.00057
陈宇聪, 谭伟锋, 戎伟鑫, 韩春春, 王耿彬. 基于大语言模型的病历生成智能体研究与设计. 2025. doi: 10.12201/bmr.202510.00038
王蕾, 刘苗, 王茜, 赵琬清, 胡佳慧, 方安. 生成式人工智能医学语料库:数据风险、合规义务及应对方案. 2025. doi: 10.12201/bmr.202512.00083
-
序号 提交日期 编号 操作 1 2026-09-22 10.12201/bmr.202611.00037V1
下载 -
-
公开评论 匿名评论 仅发给作者
引用格式
访问统计
- 阅读量:12
- 下载量: 0
- 评论数:0

登录
注册




京公网安备