摘要
数据挖掘技术是分析复杂医疗病例的重要手段,但如何选择合适的指标将中医临床经验科学化以提高医生的诊断和治疗水平、并对其进行有效监控是中医急需解决的问题.针对这个问题,提出一种医疗指标约简的方法——基于聚类的贝叶斯网络模型,通过分类及进行主特征和类特征的提取来研究中医症候分型、动态演变,为类风湿关节炎病因、病机的研究及临床医生诊断质量控制提供了重要依据.
Data mining technology is an important method to analyze the medical cases. It is important to choose the appropriate indices to make the Chinese medicine more scientific and to improve the diagnosis level by supervising the quality more effectively. This paper brings forward a new medical indices reduction methodthe Bayesian network based on clustering to solve this problem. This paper studies TCM syndrome type and dynamic evolution by classifying the data to pick-up the main characters and species characters. It provides the important evidence to RA pathogens and sickness mechanism research, and it also plays an important role in supervising diagnosis quality of the clinician.
出处
《管理科学学报》
CSSCI
北大核心
2008年第6期143-150,共8页
Journal of Management Sciences in China
基金
国家自然科学基金资助项目(70501016)
关键词
临床实验
数据挖掘
聚类
贝叶斯网络
clinic experiment
data mining
clustering
Bayesian network
