数据挖掘在肺结核疾病智能决策中的应用研究

Application of Data Mining in Intelligent Decision of Pulmonary Tuberculosis Diseases

针对单一数据挖掘方法对肺结核疾病诊断效率低、准确性不高的问题,本研究对北京市昌平区结核病防治所,北京市结核病控制研究所的1203例肺结核病人档案资料构建了电子档案,采用粗糙集和决策树结合方法建立肺结核疾病诊断模型,并对其准确性进行评估,在此基础上构建肺结核疾病诊断系统;在研究中,使用粗糙集和决策树相结合的方法进行属性约简,约简掉冗余属性57个,剩余属性22个,得到决策规则7条,模型准确率为83.46%;与未未约简的方法相比,决策规则减少128%,模型准确率基本保持不变;研究结果表明:使用该组合算法,在保证模型准确率的同时,降低了算法的时间和空间复杂性,提高了挖掘效率,为临床诊断提供了一定的借鉴。

Aiming at the problem that the low diagnostic etiiciency and low accuracy o~ the single data mmmg method for JAlagnosls OI pulmonary tuberculosis, In this study, the electronic records of 1203 cases of tuberculosis patients in Changping District City, Beijing City of Beijng and Beijing Institute of tuberculosis control and tuhereulosis control were build, Tuberculosis disease diagnosis model is built by application of rough set and decision tree method, On the basis of this, the diagnosis system of pulmonary tuberculosis was constructed. In this study, The combining method of rough set and decision tree was approached to attribute reduction, the model reduced redundant 57 attributes and remained 22 attributes, and articled 7 the decision rules. The model accuracy is 89.46%. Compared with the non reduction method, the decision rule was reduced by 128%, and the accuracy of the model remained unchanged. The research results showed that the algorithm can reduce the time and space complexity of the algorithm while ensuring the accuracy of the model, so as to improve the efficiency of the mining, and provide some references for clinical diagnosis.