الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 95 |  |  |
| | | from | 95 | | |
Abstract
Nowadays, medical imaging technologies provide the clinician with some complementary diagnostic tools, such as x-ray, computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound. The brain is one of the most complex organs of the human body, so it is a difficult task to differentiate its various components and deeply analyze them. MRI is very common for brain image analysis. MRI’s are an advanced medical imaging technique providing rich information about the human soft tissue anatomy. The goal of brain MR image segmentation is to identify the principal tissue structures in these image volumes accurately. Tissue segmentation of normal and pathological tissue types using MR images has great potential in clinical practice. Possible areas of application include the automatic and semi-automatic delineation of tumors before and after surgical or radio-surgical intervention for response assessment. Tissue segmentation is also of importance in the study of neurodegenerative diseases such as Alzheimer’s diseases. Segmentation of MR images for computer-aided diagnosis is often required as a preliminary stage. To analyze brain data to bring useful information for diagnostic, MRI must be segmented into different tissues that composing it. The main brain tissues are white matter (WM), gray matter, and cerebrospinal fluid. Moreover, regional volume calculations of these tissues may bring more useful diagnostic information. For example, calculation of the quantization of gray and white matter volumes may be of major interest in neurodegenerative disorders, such as Alzheimer disease, in movements disorders such as Parkinson or Parkinson related syndrome, in white matter metabolic or inflammatory disease, or perinatal brain damage, or in post-traumatic syndrome. Previously in many clinical studies, segmentation is still mainly manual or strongly supervised by a human expert. The level of operator supervision impacts the performance of the segmentation method in terms of time consumption, leading to infeasible procedures for large datasets. Moreover, due to the characteristic of MRI, there are mainly three considerable difficulties for segmenting MRI such as noise and intensity inhomogeneity that makes the manually segmenting result is not accurate. The main contribution of the work we present here a Multi-resolution MRI Brain Image Segmentation Based on Morphological Pyramid and Fuzzy C-mean Clustering. The proposed work consists of three stages as the Following: the Haar wavelet stage, morphological pyramid, and FCM. We compared our proposed system with some state of the art segmentation techniques on two different brain data sets. Experimental results showed that the proposed system improves the accuracy of the MRI brain image segmentation with 97.05%. Another main contribution of the work we present here MRI Brain Image Segmentation Based on Cascaded Fractional-order Darwinian Particle Swarm Optimization and Mean Shift. The proposed framework is based on two segmentation methods: Fractional-order Darwinian Particle Swarm Optimization (FODPSO) and Mean Shift segmentation (MS). In the pre-processing phase, the MRI image is filtered, and the skull is removed. In the segmentation phase, the result of FODPSO is used as the input to MS. Finally, we make a validation to the segmented image. We compared our proposed system with some state of the art segmentation techniques using brain benchmark data set. The experimental results show that the proposed system enhances the accuracy of the MRI brain image segmentation with 0.9821. In this thesis the proposed frameworks are implemented in MATLAB R2011a on a Core(TM) 2 Due 2 GHz processor and 4GB RAM, the experimental results show that the first proposed system superior to KM, EM, FCM, and KFCM_S2. Image segmented into four classes WM, GM, CSF, and background. In the second proposed framework Results indicate that the use of both segmentation methods can overcome the shortcomings of each other(FODPSO+MS), and the combination can improve the result of the classification process significantly.