الفهرس 
Chapter 1 IntroductionOnly 14 pages are availabe for public view
 |  | from | 127 |  |  |
| | | from | 127 | | |
Abstract
Human brain and spinal cord form the Central Nervous System (CNS), where all of the vital functions are ruled. These functions include movement, vision, speech, and hormones regulation. For that reason, in case of brain tumor presence, it will directly affect one or more vital human functions. Until now, there are no way to prevent brain tumors. Therefore, early detection and diagnosis of brain tumor will increase the probabilities of complete recovery and thus mortality rate will be reduced. Computer Aided Diagnosis systems (CADs) have been developed to be a powerful tool in the early detection of brain tumors by classifying and segmenting brain tumors. These systems act as a second opinion to aid the radiologists in brain tumor diagnosis enhancing. They help in processing large scale images and involve digital image processing, image analysis, and ma- chine learning. The work is divided into three parts. The first part of this work is to classify brain tumor images into their types ; in one study, classification of two brain tumor types (Benign, and Malignant) is performed. In another study, classification of three brain tumor types (meningioma, glioma, and pituitary tumor). Datasets include 3586, 3504 T1-weighted contrast-enhanced images for the first and second study respectively. Results for tumor type classification and grading show that the proposed architecture effectively detects and classify tumors images. The brain images are classified into (Benign, and Malignant) tumor with overall accuracy of 99.8% in the first study of classification. On the other hand, results of 99.6% is obtained for classified into meningioma, glioma, or pituitary tumor. Finally hybrid model with overall accuracy of 97.2%.