Texture and Shape Based Classification of Brain Tumors Using Back-Propagation Algorithm

Accuracy and efficiency are two major issues in designing CAD (Computer Aided Diagnosis) systems. Most of CAD systems are dedicated to visual feature extraction because it has been shown that visual information extracted from images can achieve similarity retrievals with high performance of effectiveness of the diagnosis, at the same time reducing the pain of the patients also. In the brain MR image, the tumor may appear clearly, but for further treatment, the physician also needs the quantification of the tumor area. The computer and image processing techniques can provide great help in analyzing the tumor area. In this paper features based on content of image were tested for analysis and classification of brain tumor using texture and shape analysis. After feature extraction back-propagation algorithm is used to classify brain tumor in to malignant & benign. MATLAB ® 7.01, its image processing toolbox and ANN toolbox have been used to implement the algorithm. The results show that texture and shape features can be effectively used for classifying brain tumor with high level of accuracy. Keywordsmedical image, shape, texture, malignant, benign, tumor.