文献类型：期刊文献

英文题名：Risk Factors Analysis and Classification on Heart Disease

作者：Luo, Jianfeng[1,2];Yan, Haifeng[1];Yuan, Yubo[2,3]

机构：[1]East China Univ Sci & Technol, Sch Business, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[3]East China Univ Sci & Technol, Ecust Medsci Res Ctr Biomed Big Data, Shanghai 200237, Peoples R China

年份：2020

卷号：24

期号：17

起止页码：13167

外文期刊名：SOFT COMPUTING

收录：;EI(收录号：20200708162189);WOS:【SCI-EXPANDED(收录号:WOS:000515989000003)】；

语种：英文

外文关键词：Canonical correlation analysis; Heart disease; Functional classification; Risk factors analysis; Group factors

摘要：In recent years, there has been a high prevalence rate of heart disease (HD) among 50-year-old people in China. It has become the first disease of old ages death. It is a very interesting and challenging work to have an effective early forecasting of the risk of HD according to the patients data. In this paper, we propose a novel method to analyze the factors with views of group features. Normalized mutual information based on entropies and information gain ratio are employed to select factors. Discriminant minimum class locality preserving canonical correlation analysis is presented to determine the effectiveness of the view of group factors. Moreover, a novel model is given to forecast the risks of New York Heart Association Functional Classification. To verify the effectiveness of the proposed method and model, we collected electronic health records of 1271 patients from 28 Chinese Level III-A hospitals in 2015. After the risk factors analysis, several results are concluded: (1) Patients with HD usually suffer from similar complications. For example, most patients with heart disease suffer from hypertension, diabetes and arrhythmia at the same time. (2) The risk forecasting has an accurate recognition rate. The risk value of the level of patients is impacted on the complications. (3) Hypertension, arrhythmia, chronic cardiac insufficiency and coronary disease are the highest concurrent diseases. There is a high reliability to have a decision of levels on the cardiac functional diseases according to the output of our proposed model.