Computer Based Advanced Approach for MRI Image Classification Using Neural Network With The Texture Features Extracted

摘要

Now a days,In order to look at organs inside the bodies Magnetic Resonance Imaging (MRI)utilizes radio waves and large magnet. This paper mainly provides an magnificent approach for automatic detection and classification is presented. Image classification,reprocessing and feature extraction are the main approaches in proposed system. In the pre-processing stage we used M3 Filtering Algorithm are applied for the removal of noise,Texture for capturing the contents of the images for indexing,feature extraction technique is applied. Classification of MRI images is carried out in the categories of Brain, Heart, Hands, Legs images using Probabilistic SVM(Support Vector Machine), PNN(probabilistic Neural Network), BPNN(Back Propagation Neural Network) algorithms. The method was applied using 400 images are collected from Andhra Hospital, Vijayawada. An overall accuracy of 92% is achieved with the MRI image classification using SVM contrast with other algorithms.