数据挖掘在胎儿心率上的规则预测研究  

Study on rule prediction of data mining in fetal heart rate

作　　者：刘秋菊 陈宋[1] 黄立勤[1] 

机构地区：[1]福州大学物理与信息工程学院,福建福州350108

出　　处：《微型机与应用》2017年第19期16-18,共3页Microcomputer & Its Applications

基　　金：国家自然科学基金重点项目资助(61471124)

摘　　要：瞬时胎心率是监测胎儿健康状态的一种重要方式。当前,监控胎儿心率是重要而复杂的任务,正确的自动化分类和规则提取是非常必要的。医疗诊断自动化系统,不仅加强医疗保健,同时也可以降低成本。设计了一个有效挖掘规则,并根据给定的参数来预测胎儿的风险水平。采用C4.5、Classification and Regression Tree(CART)、随机森林分类器来进行系统比较。该系统的性能评价由分类精度、产生规则数量构成。实验结果表明,基于随机森林分类器的系统具有高精度(99.4%)的预测胎儿健康状态的潜力,同时,产生的规则数量精简且可供于医生决策。Fetal heart rate is an important way to monitor fetal health. At present,the monitoring of fetal heart rate is an important and complex task,and correct automatic classification and rule extraction are necessary. Medical diagnostic automation systems will strengthen health care,but also reduce costs. In this study,we designed an effective rule to predict the risk level of fetus. Then,we used C4. 5,Classification and Regression Tree（ CART）,random forest classifier to evaluate the system. The performance of the system is evaluated by the classification accuracy,which produces the number of rules. The experimental results show that the system based on the random forest classifier has high accuracy（ 99. 40%） in predicting the health status of the fetus,while the number of rules produced is simplified and can be used for decision making.

关 键 词：随机森林 瞬时胎心率 规则提取 C4.5 CART 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] Q-332[自动化与计算机技术—计算机科学与技术]