| 一种基于决策树的乳腺癌计算机辅助诊断新方法 |
| |
| 引用本文： | 毛利锋,瞿海斌.一种基于决策树的乳腺癌计算机辅助诊断新方法[J].江南大学学报(自然科学版),2004,3(3):227-229. |
| |
| 作者姓名： | 毛利锋 瞿海斌 |
| |
| 作者单位： | 浙江大学,药物信息学研究所,浙江,杭州,310027 |
| |
| 基金项目： | 国家自然科学基金项目(30000218)资助课题，国家中医药管理局科研基金重点项目(2000 J Z 03)资助课题. |
| |
| 摘 要： | 选取500例乳腺癌病例为数据样本，每个样本由9个细针吸取细胞学指标组成，将样本随机分为训练集和测试集，然后利用决策树方法从训练集中学习得到诊断模型，经测试集测试．结果表明决策树的诊断准确率高迭97．33％，灵敏度和特异度分别为98．28％和96．74％．因此，决策树是一种简便可行的计算机辅助诊断方法．
|
| 关 键 词： | 决策树 乳腺癌 细针吸取细胞学 数据挖掘 诊断 |
| 文章编号： | 1671-7147(2004)03-0227-03 |
A New Computer-Aided Method for Diagnosis of Breast Cancer Based on Decision Tree |
| |
| MAO Li-feng,QU Hai-bin.A New Computer-Aided Method for Diagnosis of Breast Cancer Based on Decision Tree[J].Journal of Southern Yangtze University:Natural Science Edition,2004,3(3):227-229. |
| |
| Authors: | MAO Li-feng QU Hai-bin |
| |
| Abstract: | In this research the efficacy and prospect of applying decision tree method in assisting fine needle aspiration cytology (FNAC) for breast cancer diagnosis is evaluated. Data from 500 breast cancer patient records comprised of 9 FNAC variables in each sample are randomly divided into training set and testing set. The decision tree method is used to construct a classification model for the training data. The results show that the model achieves up to 97.33% of classification accuracy, 98.28% of sensitivity of and 96.74% of specificity. Therefore, the decision tree method is a simple and useful tool for computer-aided diagnosis. |
| |
| Keywords: | decision tree breast cancer fine needle aspiration cytology data mining diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
