Weiner Filter based Hough Transform and Wavelet feature extraction with Neural Network for Classifying Brain Tumor

The detection of MRI brain abnormality using CAD system is a perplexing task for radiologists. To overcome this difficulty a new CAD system approach is proposed for detecting the brain abnormality like tumor. Initially the input image is preprocessed using Weiner filter and segmented using Hough Transform. After that, the structures are extracted using Wavelet Transform. Finally the image is classified using Probabilistic Neural Network (PNN) from the extracted features. PNN classifier classifies the MRI brain tumor into two, namely benign and malignant. The valuation of the proposed work is done by associating it with the prevailing neural network methods. From the obtained results it is concluded that proposed CAD approach is giving better results than the existing methods by giving less error rate.