Brain MRI segmentation using the mixture of FCM and RBF neural network

Author

Rostami, Maryam Talebi ; Ghaderi, Reaza ; Ezoji, Mehdi ; Ghasemi, Javad

Author_Institution

Electr. & Comput. Dept., Babol Univ. of Technol., Babol, Iran

fYear

2013

Firstpage

425

Lastpage

429

Abstract

One of the most commonly used methods for Magnetic Resonance Imaging (MRI) segmentation is Fuzzy C-Means (FCM). This method in comparison with other methods preserves more information of the images. Because of using the intensity of pixels as a key feature for clustering, Standard FCM is sensitive to noise. In this study in addition to intensity, mean of neighbourhood of pixels and largest singular value of neighbourhood of pixels are used as features. Also a method for segmenting MRI images is presented which uses both FCM and Radial Basis Function (RBF) neural network and partly decreases the limitation of standard FCM.

Keywords

biomedical MRI; fuzzy set theory; image segmentation; medical image processing; pattern clustering; radial basis function networks; FCM; RBF neural network; brain MRI segmentation; fuzzy c-means; image formation; magnetic resonance imaging; pixel intensity; pixels neighbourhood; radial basis function neural network; Clustering algorithms; Educational institutions; Image segmentation; Indexes; Magnetic resonance imaging; Neural networks; Noise; FCM; MRI segmentation; RBF neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Vision and Image Processing (MVIP), 2013 8th Iranian Conference on

Conference_Location

Zanjan

ISSN

2166-6776

Print_ISBN

978-1-4673-6182-8

Type

conf

DOI

10.1109/IranianMVIP.2013.6780023

Filename

6780023