首页 | 本学科首页   官方微博 | 高级检索  
     检索      

结直肠癌肿瘤标志物多分割点设置算法
引用本文:施京华,苏强,顾平,黄钢,朱岩.结直肠癌肿瘤标志物多分割点设置算法[J].清华大学学报(自然科学版),2009(5).
作者姓名:施京华  苏强  顾平  黄钢  朱岩
作者单位:上海交通大学工业工程与物流工程系;上海交通大学医学院附属仁济医院核医学科;清华大学经济管理学院;
基金项目:国家自然科学基金资助项目(70672077);;上海交通大学医工(理)交叉研究基金重点项目(YG2007ZD10)
摘    要:为了提高诊断正确率,大量医学研究集中于探讨肿瘤标志物正常值与非正常值的合适划分,即确定分割点。为此,该文提出了一种肿瘤标志物多分割点设置算法。通过正交实验和缩小搜索空间的方法,进一步提高了诊断正确率和优化收敛速度。在结直肠癌实际诊断数据中的应用效果表明,该算法能有效地提高诊断正确率,其诊断结果优于医学上的单分割点的诊断结果。

关 键 词:粗糙集  遗传算法  结直肠癌  肿瘤标志物  诊断正确率  正交实验  

Multi-cutoff level setting algorithm for various tumors markers in colorectal cancer
SHI Jinghua,SU Qiang,GU Ping,HUANG Gang,ZHU Yan.Multi-cutoff level setting algorithm for various tumors markers in colorectal cancer[J].Journal of Tsinghua University(Science and Technology),2009(5).
Authors:SHI Jinghua  SU Qiang  GU Ping  HUANG Gang  ZHU Yan
Institution:1.Department of Industrial Engineering and Logistic Management;Shanghai Jiaotong University;Shanghai 200240;China;2.Department of Nuclear Medicine;Renji Hospital;School of Medicine;Shanghai 200001;3.School of Economics and Management;Tsinghua University;Beijing 100084;China
Abstract:There have been many studies to determine the best cutoff levels for tumor markers to differentiate between normal and abnormal values.This paper presents a multi-cutoff level setting algorithm for various tumors markers to improve diagnostic accuracy.The accuracy and convergence of the algorithm are further improved using an orthogonal experiment and limiting the search space.Experiments demonstrate that this algorithm effectively improves the diagnostic accuracy for colorectal cancer by various tumor mark...
Keywords:rough set theory  genetic algorithm  colorectal cancer  tumors markers  diagnostic accuracy  orthogonal experiment  
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号