期刊文献+

乳腺癌5年生存预测模型的建立及验证 被引量:1

下载PDF
导出
摘要 目的建立乳腺癌5年生存预测模型并加以验证。方法利用数据挖掘技术进行数据预处理,分别建立乳腺癌5年生存预测Logistic模型和Nave Bayes模型,10折分层交叉验证法对模型性能进行测试。结果Logistic模型准确度为0.8855±0.0015,ROC曲线下面积(AUC_(Roc))为0.8060±0.0227;Nave Bayes模型准确度为0.8422±0.0026,AUC_(Roc)为0.7090±0.0514。结论对于乳腺癌的生存预测,Logistic模型优于Nave Bayes模型。
出处 《上海交通大学学报(医学版)》 CAS CSCD 北大核心 2008年第11期1481-1483,共3页 Journal of Shanghai Jiao tong University:Medical Science
  • 相关文献

参考文献9

  • 1Li L, Tang H, Wu Z, et al. Data mining techniques for cancer detection using serum proteomic profiling [ J ]. Artif Intell Med, 2004,32(2) :71 -83.
  • 2NCI. SEER Data, 1973 - 2004 [ dB/OL]. http://seer. cancer. gov.
  • 3Han JW. Data mining concepts and techniques [ M ]. 2nd ed. Beijing: China Machine Press, 2006:237-238.
  • 4Vergouwe Y, Steyerberg EW, Eijkemans MJ, et al. Substantial effective sample sizes were required for external validation studies of predictive logistic regression models [ J ]. J Clin Epidemiol, 2005, 58(5) :475 -483.
  • 5Mozina M, Demsar J, Kattan MW, et al. Nomograms for visualization of naive bayesian classifier [ C ]//Proceedings of the principles practice of knowledge discovery in databases (PKDD-04). Pisa, 2004, 337 -348.
  • 6Brumback LC, Pepe MS, Alonzo TA. Using the ROC curve for gauging treatment effect in clinical trials[ J]. Stat Med, 2006, 25 (4) : 575 -590.
  • 7Delen D. Walker G, Kadam A. Predicting breast cancer survivability: a comparison of three data mining methods [ J]. Artif Intell Med, 2005, 34(2) :113 -27.
  • 8NCI. 5 - year Relative Survival Rates [ DB/OL]. http ://seer. cancer. gov.
  • 9Abdelghani B,Erhan G. Predicting breast cancer survivability using data mining techniques[ C]//Chandrika K,Michael B. Proceedings of the 6th SIAM International Conference on Scientific Data Mining. Maryland: SIAM,2006:1-4.

同被引文献4

引证文献1

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部