Classification of magnetic resonance images using support vector machines

Sookpotharom, Supot; Tungjitkusolmun, Supan; Airphaiboon, Surapan; Sangworasil, Manas

Classification of magnetic resonance images using support vector machines

Author(s)

Sookpotharom, Supot

Tungjitkusolmun, Supan

Airphaiboon, Surapan

Sangworasil, Manas

Date Issued

December 1, 2001

Type

Conference Paper

Abstract

The classification of medical images obtained from magnetic resonance imaging (MRI) is an important step in the visualization of soft issue in the human body. MRI is a multidimensional technique as it provides information about three tissue dependent parameters. This paper presents the potential of Support Vector Machines (SVMs) technique for the supervised classification of MRI images. The SVMs approach was originally developed for binary classification problems. In this paper SVM architectures for multi-class classification are used, in particular we consider binary trees of SVMs to solve the multiclass of MR brain images. The experiments using the SVMs technique presented in this paper performed the quality and correctly position of the internal organ.

Citation

International Symposium on IC Technology Systems and Applications, 9, 488-491, 2001