主管单位:中华人民共和国
国家卫生健康委员会
主办单位:
总编辑:杨秋
编辑部主任:吴翔宇
邮发代号:80-528
定价:28.00元
全年:336.00元
Email:zgyy8888@163.com
电话(传真):010-64428528;
010-64456116(总编室)
英文作者:Gong Fan1 Chen Qingguang2 Lu Hao1
单位:1上海中医药大学附属曙光医院内分泌科,上海201203;2上海中医药大学附属曙光医院糖尿病研究所,上海201203
英文单位:1Department of Endocrinology Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine Shanghai 201203 China; 2Diabetes Institute of Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine Shanghai 201203 China
关键词:血糖管理;知识图谱;人工智能
英文关键词:Bloodglucosemanagement;Knowledgegraph;Artificialintelligence
血糖管理是糖尿病治疗的关键点,临床医师面对巨量医学信息时需要实现治疗决策的实时更新而不得不面临着信息超载问题。如何从海量数据中发现隐含的知识和规律是人工智能算法的优势。然而医学数据的多源分布导致该决策无法满足规范化及个体化原则。知识图谱是目前解决数据融合及信息超载最有效的方法之一,通过面向特定领域形式化的对共享概念体系的详细说明,从数据中通过信息抽取的方法获取知识,并对多个知识进行关联、合并,使其成为有机整体以提供更全面的知识。通过知识图谱的应用可弥补数据稀及不完整所致推荐算法性能的不足,提高血糖管理决策质量,为血糖管理提供精准、规范的指导。
Blood glucose management is the key point in the treatment of diabetes mellitus. Clinicians have to face the problem of information overload because they need to update treatment decisions in real time in facing huge amount of medical information. How to find hidden knowledge and rules from massive data is the advantage of artificial intelligence algorithm. However, due to the multi-source distribution of medical data, the decision making can′t meet the principles of standardization and individualization. At present, knowledge graph is one of the most effective methods to solve data fusion and information overload. Through a clear and detailed description of the sharing concept system in a specific domain, knowledge is obtained from data by the method of information extraction, and multi-knowledge are associated and merged to make it integrity to provide more comprehensive knowledge. Through the application of knowledge graph, we can make up for the deficiency of recommendation algorithm caused by the scarcity and incompleteness of data, improve the decision quality of antihyperglycemic decision, and provide accurate and standardized guidance for blood glucose management.
copyright
地址:北京市朝阳区安贞路2号首都医科大学附属北京安贞医院北楼二层
电话:010-64456116 传真:010-64428528 邮编:100029 Email: zgyy8888@163.com
网址: 京ICP备2020043099号-3
当您在使用本网站投稿遇到困难时,请直接将稿件投送到编辑部邮箱zgyy8888@163.com。