RAMAKRISHNAN and SELVAN: CLASSIFICATION OF BRAIN TISSUES

This paper describes a new approach for the classification of brain tissues into White Matter, Gray Matter, Cerebral Spinal Fluid, Glial Matter, Connective and MS lesion in multiple sclerosis. The proposed approach employs singular value decomposition on multiwavelet transformed images. Single level multiwavelet transformation decomposes images into 16 subbands, and each subband represents the image in a specific time frequency plane. Singular value decomposition is then employed on the subband coefficient matrices. Lower singular values are affected more by noise than higher singular values, and hence only higher singular values are used to classify textures in the presence of noise. The probability density function of the selected singular values is then modeled as an exponential distribution, and the model parameter for the distribution is estimated using the maximum likelihood estimation technique. The model parameters, one for each subband are used as features for the classification. The classification is carried out using Weighted Probabilistic Neural Networks (WPNNs). Compared to conventional probabilistic neural networks, WPNNs include weighting factors between pattern layer and summation layer of the conventional PNNs. Experiments have been carried out using data sets composed of three modalities of brain MR images, namely T1 and T2 relaxation times and proton density (PD) weighted MR images. The performance of the algorithm is analyzed in terms of classification rate at various noise levels and intensity non-uniformity levels. The experimental results demonstrate that the proposed algorithm improves the classification rate in the presence of noise and in the presence of intensity non-uniformity levels.