TY - GEN
T1 - Discriminant method for severity of glandular tumor by support vector machine
AU - Suzuki, Ayako
AU - Tanaka, Toshiyuki
PY - 2008
Y1 - 2008
N2 - In this study, glandular tumor images are classified automatically by the support vector machine (SVM) in order to make up for a fault of discriminant analysis, Mahalanobis' generalized distance which was used in recent studies. The fault of Mahalanobis' generalized distance is the problem, that is to say, the Curse of Dimensionality. To avoid this problem, we used the support vector machine (SVM) as the discriminant analysis, used the prostate images as glandular tumor images ,and examined the effectiveness of this system.
AB - In this study, glandular tumor images are classified automatically by the support vector machine (SVM) in order to make up for a fault of discriminant analysis, Mahalanobis' generalized distance which was used in recent studies. The fault of Mahalanobis' generalized distance is the problem, that is to say, the Curse of Dimensionality. To avoid this problem, we used the support vector machine (SVM) as the discriminant analysis, used the prostate images as glandular tumor images ,and examined the effectiveness of this system.
KW - Discriminant analysis
KW - Support vector machine
KW - Texture analysis
UR - http://www.scopus.com/inward/record.url?scp=56749170476&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=56749170476&partnerID=8YFLogxK
U2 - 10.1109/SICE.2008.4655197
DO - 10.1109/SICE.2008.4655197
M3 - Conference contribution
AN - SCOPUS:56749170476
SN - 9784907764296
T3 - Proceedings of the SICE Annual Conference
SP - 3101
EP - 3104
BT - Proceedings of SICE Annual Conference 2008 - International Conference on Instrumentation, Control and Information Technology
T2 - SICE Annual Conference 2008 - International Conference on Instrumentation, Control and Information Technology
Y2 - 20 August 2008 through 22 August 2008
ER -