Detection and Identification of Brain Tumor in Brain MR Images Using Fuzzy C-Means Segmentation

Abstract: In this paper a novel brain MR image segmentation method is presented based on self organizing map (SOM) neural network. An accurate segmentation of brain tissues provides a way to identify many brain disorders. This paper presents unsupervised approaches for brain image segmentation. The proposed method consists of four stages. Initially an anisotropic diffusion filtering is used as a pre-processing step to eliminate bias field and random noise. Then Stationary wavelet transform (SWT) is applied to the images to obtain multi-resolution information for distinguishing different tissues. Statistical information of the different tissues is extracted by applying spatial filtering to the coefficients of SWT. These features are combined together with the raw wavelet transform coefficients to obtain a feature vector. This feature vector is applied to the SOM network. SOM is used to segment images in a competitive unsupervised training methodology. The output images show that the proposed method generated more segmented details of the input images.

Keywords: Anisotropic diffusion filter, Stationary Wavelet Transform, Feature Vector, Self Organizing Map.