杨金平, 何方劲, 徐海峰, 杨鹏程. 基于LLM的新疆多语种医院智能客服系统构建. 2026. biomedRxiv.202609.00005
基于LLM的新疆多语种医院智能客服系统构建
通讯作者: 杨鹏程, yang_pc@yeah.net
DOI:10.12201/bmr.202609.00005
Construction of a Multilingual Hospital Intelligent Customer Service System for Xinjiang Based on Large Language Models
Corresponding author: Pengcheng Yang, yang_pc@yeah.net
-
摘要:目的/意义针对新疆多民族聚居区语言障碍导致的就医流程复杂、导诊效率低等问题,构建高效、公平的本地化智能客服系统,提升少数民族患者就医体验与服务可及性。方法/过程基于Qwen3大模型与RAG技术,构建汉、维、哈三语交互系统,集成多语种语音交互与 RAG 检索,采用轻量化部署以适应边疆网络条件,并在新疆某三甲医院试点三个月。结果/结论预约挂号成功率由60%提升至92%,平均响应时间从8min缩短至15s,多语种用户满意度达91%。该系统有效缓解语言壁垒,显著提升医院服务效率与患者满意度,为人工智能赋能民族地区医疗公平提供可复制的技术范式。
Abstract: Objective/Significance: To address the issues of complex medical procedures and low triage efficiency caused by language barriers in multi-ethnic regions of Xinjiang, this study develops an efficient and equitable localized intelligent customer service system, aiming to improve healthcare accessibility and the medical experience for ethnic minority patients.Methods/Process: Based on the Qwen3 large language model and retrieval-augmented generation (RAG) technology, a trilingual interactive system (Chinese, Uyghur, and Kazakh) was constructed, integrating multilingual voice interaction and RAG-based knowledge retrieval. The system adopts lightweight deployment strategies to adapt to the limited network conditions in border areas and was piloted for three months at a tertiary hospital in Xinjiang.Results/Conclusion: The appointment registration success rate increased from 60% to 92%, the average response time decreased from 8 minutes to 15 seconds, and user satisfaction among multilingual patients reached 91%. The system effectively alleviates language barriers, significantly enhances hospital service efficiency and patient satisfaction, and provides a replicable technical paradigm for leveraging artificial intelligence to promote healthcare equity in ethnic regions.
Key words: Slarge language model; multilingual; hospital intelligent customer service提交时间:2026-09-03
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
-
申艳妮, 惠婷. 基于大语言模型的医疗机构图像智能化归档系统设计. 2026. doi: 10.12201/bmr.202602.00047
吴宏, 胡军, 陈尔真, 董晨杰, 李建华, 叶琪. 基于大模型的医疗质控应用系统构建. 2025. doi: 10.12201/bmr.202503.00004
梁乐然, 阎誉豪, 唐跃豪, 练洢彤, 徐倩, 杨猛, 伍家亨, 陈振虎, 刘秀峰. 基于知识图谱和大语言模型的中医药治疗帕金森病智能问答研究. 2026. doi: 10.12201/bmr.202604.00033
赖书兰, 邓盼盼, 李军莲. 大语言模型在生物医学学术服务平台的应用分析及启示. 2026. doi: 10.12201/bmr.202604.00083
刘会, 姚宽达, 刘湘闽, 胡佳慧, 王茜, 方安. 基于知识图谱增强的大语言模型预检分诊框架研究. 2026. doi: 10.12201/bmr.202608.00014
杨亚洲, 郑石林, 周小毅, 章施静, 赵坚. 基于知识图谱的HIV防治大模型RAG问答系统构建研究. 2025. doi: 10.12201/bmr.202511.00061
胡瑾, 郭安辉, 杨冰晴, 赵士洁, 朱卫国. 从数据到决策:医院临床科研模型的智能化构建与应用. 2026. doi: 10.12201/bmr.202604.00035
蒋宏建, 陈健超. 基于医院大数据的智能就诊路径实时推荐系统探索与设计. 2024. doi: 10.12201/bmr.202408.00055
施呈昊, 屠馨怡, 史佳伟, 陈红爽, 王琴潞, 邹海欧. 大语言模型在临床实践中应用的范围综述. 2024. doi: 10.12201/bmr.202406.00001
李日丽, 潘嘉明, 容世强, 孙晓翠, 易法令. 知识图谱与大语言模型融合在糖尿病健康管理中的应用进展. 2025. doi: 10.12201/bmr.202508.00035
-
序号 提交日期 编号 操作 1 2026-04-22 10.12201/bmr.202609.00005V1
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:23
- 下载量: 0
- 评论数:0

登录
注册




京公网安备