文献类型：会议论文

英文题名：RISK ESTIMATION FOR AGE OF CHRONIC DISEASES

作者：Zhang, Huan-Huan[1];Cai, Zheng[1];Yuan, Yu-Bo[1];Wang, Ying-Jie[2];Gao, Ju[2]

机构：[1]East China Univ Sci & Technol, Shanghai 200237, Peoples R China;[2]Shanghai Univ Tradit Chinese Med, ShuGuang Hosp, Shanghai 201203, Peoples R China

会议论文集：International Conference on Machine Learning and Cybernetics (ICMLC)

会议日期：JUL 09-12, 2017

会议地点：Ningbo, PEOPLES R CHINA

语种：英文

外文关键词：Chronic diseases; Double well basic function; Least squares fitting

摘要：Chronic disease is a class of diseases which seriously threatens the health of our residents, it has become a major public health problem affecting the economic and social development of the country. And these chronic diseases are mostly old age diseases, age is an important risk factor. In this paper, we construct four double well basic functions, and aim at minimizing the segmental fitting error, we finally get the optimal segmentation point and construct the fitting function. We have tested on the age data set of 12 chronic diseases. The results show that the proposed model is better than the least squares fitting models. Moreover, the accuracy of this model to estimate the age risk of chronic disease is significantly improved. For every chronic disease, we get three key ages which have an interesting reference to clinical analysis.