Paper Info
Title
Discover High-Risk Factor Combinations Using Bayesian Network From National Screening Data In China
Abstract
Stroke, characterized by high incidence, high prevalence and high mortality, has a serious impact on the health of the residents, and brings heavy burden to the family and the society in China. It is proved that early screening is an important and effective method in stroke prevention and control. Based on the 2012 screening data from China stroke screening and intervention program, we manually define a network and use Bayesian network model to study the network parameter and stroke probability with risk factors in order to explore the rationality of the screening standard proposed by the program mentioned above. We find that it is not reasonable enough to simply consider those with three risk factors as of high-risk for stroke without distinguishing the types and combinations of risk factors. Some two-factors combinations are with higher stroke probability than three-factors combinations. Therefore, in the screening, some new criteria should be added to avoid missing person with high risk of stroke, and a threshold is needed to remove low-risk combinations in order to save economic costs.
Year
Venue
Keywords
2017
2017 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM)
Stroke, National screening data, Bayesian network, High-risk factors
Field
DocType
ISSN
Actuarial science,Rationality,Network parameter,Computer science,China,Stroke,Bayesian network,Artificial intelligence,Economic cost,Machine learning,Risk factor
Conference
2156-1125
Citations 
PageRank 
References 
0
0.34
0
Authors
4
Name
Order
Citations
PageRank
Xuemeng Li102.37
Jinghui Yu201.69
Mei Li3211.53
Zhao Dongsheng406.42