Brain MRI Classification-An Evaluation on PSONN and PSOSVM

Author(s):

Dhanya. G. S , Anna University-Chennai, Rajas Engineering College, Tirunelveli, Tamil Nadu, India.; Dr. R. Joshua Samuel Raj, Anna University-Chennai, Rajas Engineering College, Tirunelveli, Tamil Nadu, India.; Sam Silva. A, Anna University-Chennai, Rajas Engineering College, Tirunelveli, Tamil Nadu, India.

Keywords:

ANN, classification, MRI, segmentation, SVM

Abstract:

Advancements in technology produce huge amounts of data in various fields, increasing the need for efficient and effective data mining tools to uncover the information contained implicitly in the data. Such a voluminous store of data of diverse characteristics is mostly stored and made available in digitized form.Magnetic Resonance Imaging (MRI) plays a vital role in the diagnosis and treatment of diseases. Defect detection in Magnetic Resonance (MR) brain images is a tough task. The difficulty in brain image analysis is mainly due to the requirement of detection techniques with high accuracy within quick convergence time. Numerous automated techniques are developed to prevail over this drawback. Among these techniques, Artificial Neural Networks (ANN) and support Vector Machine (SVM) techniques are found to be highly efficient in terms of the performance measures. But, the major factor is that the merits are not simultaneously available in the same ANN or SVM technique. In this research work, several ANN and SVM techniques are studied for image segmentation and classification process. The authors have selected different features even if the method of segmentation is same. Based on that we have reached a conclusion that each technique has its own significance, but the features extracted from the segmented image play a distinct role in enhanced performance.

Other Details:

Manuscript Id	:	NCTTP013
Publication Date	:	06/05/2017
Page(s)	:	48-52

Download Article