医学数据挖掘系统研究——糖尿病并发症流行病学知识发现  

Medical Data Mining System Research Based on Computer——Epidemiological Knowledge Discovery of Diabetes Syndromes

作　　者：余辉[1] 张力新[1] 刘文耀[1] 黄志勇[1] 

机构地区：[1]天津大学生物医学工程系,天津300072

出　　处：《计算机工程与应用》2006年第18期229-232,共4页Computer Engineering and Applications

摘　　要：针对流行病学研究的特点,论文提出计算机辅助医学数据挖掘系统构架,以糖尿病并发症为研究实例,探讨医学数据的冗余性消除、规范化储存、知识归纳及可视化表达等问题。以天津总医院3022例普查数据为研究对象,尝试解决用计算机实现糖尿病并发症这类定性数据的定量化数据挖掘和知识发现。通过对于43种并发症的定性数据挖掘,可以发现诸如高血脂、冠心病、高血压、脑血管病等具有明显并发倾向的知识规则18条。同时,采用知识树方式和决策树等方法实现知识规则的可视化表达。基于数据挖掘和知识发现计算机辅助医学数据挖掘系统能够对现有病历数据库中数据进行自动分析并且提供有价值医学知识,特别适合流行病学分析和全民健康评估,因此与社区医疗和医院HIS系统结合是未来一个非常现实的发展方向。In this paper,a systematic architecture of medical data mining based on computer is provided for epidemiological analysis.Diabetes syndromes are used to discuss redundancy elimination,normalized storage,knowledge induction and visual expression of medical data,3 022 pieces of census records from Tianjin General Hospital are researched to find the solution of quantitative mining from qualitative data and knowledge discovery.From the qualitative data mining of 43 kinds of diabetes syndromes,we find 18 knowledge rules with significant statistical meaning on concurrency relation,such as hyperlipoidemia,coronary disease,hypertension,cerebrovascular disease.And knowledge tree is an effective visual expression method to show the rules generated from the above system.Medical analysis system based on compute data mining and knowledge discovery can generate effective knowledge rules from medical record database,which is especially useful to epidemiological analysis and national health survey.So how to cooperate with co mmunity medical care and hospital information system in the near future is practically significant.

关 键 词：Apriori模型 决策树 数据挖掘 知识发现糖尿病并发症 流行病学 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]