多变量数据分析模型在无症状颈动脉狭窄检测中的应用  

Application of Multivariable Data Analysis Model in Detection of Asymptomatic Carotid Artery Stenosis

作　　者：陈曦[1] 

机构地区：[1]长江师范学院数学与计算机学院,重庆408100

出　　处：《科学技术与工程》2013年第27期8100-8105,共6页Science Technology and Engineering

摘　　要：无症状颈动脉狭窄是引起中风的主要原因。而无症状颈动脉狭窄的危险因素包括高血压、心脏病发病率、吸烟、糖尿病、久坐不动等。为了明确无症状颈动脉狭窄的重要影响因素,利用多变量数据分析技术研究探讨了无症状颈动脉狭窄检测中可能存在的规则和关系。通过采用遗传算法、逻辑回归算法及卡方检验,对372位病人的临床特征进行了检测分析。实验结果表明,高血压、冠状动脉疾病是诱发无症状颈动脉狭窄最主要的两个因素,相比逻辑回归、卡方检测,遗传算法产生的简单规则更适用于医生的日常诊断。Asymptomatic carotid stenosis, one of the etiological factors for stroke, has several risk factors such as hypertension, cardiac morbidity, smoking, diabetes, and physical inactivity. To determine important factors of asymptomatic carotid stenosis, multivariable data analysis technology were used to explore rules and relationships that might be used to detect possible asymptomatic carotid stenosis. Genetic Algorithms （GA）, Logistic Regression （LR） algorithm, and Chi-square tests were also applied to detect and analyze clinical features of 372 patients. Experimental results indicate that high blood pressure and coronary artery disease were two most important factors which can cause asymptomatic carotid artery stenosis. Simple rules generated by GA can be applied to the doctor＇s diagnosis comparing with LR and Chisquare tests.

关 键 词：无症状颈动脉狭窄 多变量数据分析 数据挖掘 遗传算法 逻辑回归算法 卡方检验 

分 类 号：R331.33[医药卫生—人体生理学] TP311[医药卫生—基础医学]