面向普适健康监测的心脏病分析方法  

Analysis method of heart disease for pervasive health monitoring

作　　者：朱雪彤[1] 钱秀娟 孙薇薇[1] 侯宪春[1] 陈广生 王永利[4] 

机构地区：[1]佳木斯大学理学院,黑龙江佳木斯154007 [2]中移(杭州)信息技术有限公司,杭州310012 [3]佳木斯发电厂燃料分厂,黑龙江佳木斯154007 [4]南京理工大学计算机科学与工程学院,南京210094

出　　处：《计算机工程与应用》2018年第1期159-165,203,共8页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61170035);江苏省科技成果转化专项资金(No.BA2013047);江苏省六大人才高峰项目(No.2014WLW-004);兵科院预研项目(No.62201070151);中央高校基本科研业务费专项资金(No.30916011328)

摘　　要：为了提高普适健康监测服务的病人护理质量,采用贝叶斯网络对心脏病数据进行及时准确的分析,提出从样本数据集中学习网络节点顺序的方法,克服了传统算法需要领域专家给定网络中节点顺序的限制;另外,引入了并行优化方法进一步提高在大数据量情况下建立心脏病诊断分析模型的速度。实验证明提出的方法在一定程度上提高了模型分析的准确率,并且缩短了建模的时间。In order to promote the health-care quality of pervasive health monitoring service, this paper uses Bayesian networks to analyze heart disease, and presents a method to conform the sequence of network nodes from sample dataset.This method overcomes the limitation that traditional algorithms which require experts of medical field give the sequence of network nodes. In addition, in order to shorten the analysis time, a parallel optimization technique is adopted to accelerate the establishment of HD diagnosis model over large amounts of data. Experiments show that the proposed method can improve the accuracy of the modeling and shorten the modeling time to some extent.

关 键 词：普适健康监测 心脏病分析 贝叶斯网络 并行优化 

分 类 号：R541[医药卫生—心血管疾病] TP18[医药卫生—内科学]