Feature selection for computer-aided polyp detection using genetic algorithms

被引:23
作者
Miller, MT [1 ]
Jerebko, AK [1 ]
Malley, JD [1 ]
Summers, RM [1 ]
机构
[1] NIH, Ctr Clin, Dept Diagnost Radiol, Bethesda, MD 20892 USA
来源
MEDICAL IMAGING 2003: PHYSIOLOGY AND FUNCTION: METHODS, SYSTEMS, AND APPLICATIONS | 2003年 / 5031卷
关键词
genetic algorithms; support vector machines; feature selection; forward stepwise search; computer aided diagnosis; virtual colonoscopy;
D O I
10.1117/12.485796
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
To improve computer aided diagnosis (CAD) for CT colonography we designed a hybrid classification scheme that uses a committee of support vector machines (SVMs) combined with a genetic algorithm (GA) for variable selection. The genetic algorithm selects subsets of four features, which are later combined to form a committee, with majority vote for classification across the base classifiers. Cross validation was used to predict the accuracy (sensitivity, specificity, and combined accuracy) of each base classifier SVM. As a comparison for GA, we analyzed a popular approach to feature selection called forward stepwise search (FSS). We conclude that genetic algorithms are effective in comparison to the forward search procedure when used in conjunction with a committee of support vector machine classifiers for the purpose of colonic polyp identification.
引用
收藏
页码:102 / 110
页数:9
相关论文
共 20 条
[1]   FAST GENETIC SELECTION OF FEATURES FOR NEURAL NETWORK CLASSIFIERS [J].
BRILL, FZ ;
BROWN, DE ;
MARTIN, WN .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (02) :324-328
[2]  
CANTUPAZ E, FEATURE SUBSET SELEC
[3]   Statistical analysis of the parameters of a neuro-genetic algorithm [J].
Castillo-Valdivieso, PA ;
Merelo, JJ ;
Prieto, A ;
Rojas, I ;
Romero, G .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (06) :1374-1394
[4]  
Cristianini N., 2000, INTRO SUPPORT VECTOR, DOI [10.1017/CBO9780511801389, DOI 10.1017/CBO9780511801389]
[5]   Approximate statistical tests for comparing supervised classification learning algorithms [J].
Dietterich, TG .
NEURAL COMPUTATION, 1998, 10 (07) :1895-1923
[6]   Feature selection for optimized skin tumor recognition using genetic algorithms [J].
Handels, H ;
Ross, T ;
Kreusch, J ;
Wolff, HH ;
Pöppl, SJ .
ARTIFICIAL INTELLIGENCE IN MEDICINE, 1999, 16 (03) :283-297
[7]  
Hastie T, 2008, The elements of statistical learning, Vsecond, DOI DOI 10.1007/978-0-387-21606-5
[8]   Design of an optimal nearest neighbor classifier using an intelligent genetic algorithm [J].
Ho, SY ;
Liu, CC ;
Liu, S .
PATTERN RECOGNITION LETTERS, 2002, 23 (13) :1495-1503
[9]   Feature subset selection for classification of histological images [J].
Jelonek, J ;
Stefanowski, J .
ARTIFICIAL INTELLIGENCE IN MEDICINE, 1997, 9 (03) :227-239
[10]  
JEREBKO AK, 2003, IN PRESS ACAD RADIOL