基于AI的医疗信息可视化构建与效果评价  

Visualization Construction and Effect Evaluation of AI-based Medical Information

作　　者：肖金华 XIAO Jinhua(Department of Library Information Literacy,Guangdong Pharmaceutical University,Guangzhou 510000,Guangdong,China)

机构地区：[1]广东药科大学图书馆信息素养教研室,广东广州510000

出　　处：《医学信息》2025年第4期44-53,共10页Journal of Medical Information

基　　金：广东省医学科学技术研究基金项目(编号:B2021362)。

摘　　要：随着人工智能(AI)的日渐普及,其对医疗信息可视化构建原理与效果评价标准将产生巨大影响。本文通过对以往五年Google学术搜索和Web of Science等数据库收集的文献进行全面回顾和系统梳理,发现基于AI的医疗信息可视化构建应加强用户需求分析和个性化定制,确保信息呈现方式和内容符合用户需求。同时,应注意模型算法的选择和优化,提高准确性和可解释性。在效果评估方面,应重点关注临床影响、患者结果和可用性评估,并加强与临床医生、患者等用户的交流与反馈。As artificial intelligence(AI)becomes more and more popular,it will have a huge impact on the construction principles and effect evaluation standards of medical information visualization.Through a comprehensive review and systematic combing of literature collected from databases such as Google Scholar and Web of Science in the past five years,the study found that the construction of medical information visualization based on AI should strengthen user demand analysis and personalized customization to ensure that the information presentation method and content meet user needs.At the same time,attention should be paid to the selection and optimization of model algorithms to improve accuracy and interpretability.In terms of effect evaluation,focus should be placed on clinical impact,patient outcomes and usability evaluation,and communication and feedback with users such as clinicians and patients should be strengthened.

关 键 词：人工智能 医疗信息可视化 构建原则 效果评估 ChatGPT 

分 类 号：R-05[医药卫生] TP393.09[自动化与计算机技术—计算机应用技术]