ALGORITHMS ON MR IMAGING AND RF FIELD MAPS OF BIRDCAGE RESONATORS

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" ALGORITHMS ON MR IMAGING AND RF FIELD MAPS OF BIRDCAGE RESONATORS "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151646
Doc. No	:	ET23438
Main Entry	:	Abdelhamid Kadik
Title Proper	:	ALGORITHMS ON MR IMAGING AND RF FIELD MAPS OF BIRDCAGE RESONATORS
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	A process for classification of anatomical tissues in Magnetic Resonance (MR) images ispresented. The process uses a contextual approach for tissue classification based on the factthat the pixels in an image can be estimated from their local surrounding environment context.i.e. from the information of adjacent pixels. This procedure allows us to classify an arbitrarynumber of tissues in an interactive or operator independent way. Opentor independentalgorithms were developed for Image Segmentation of MR images for white and gray matter.ventricles and pathology. The algorithm is a multiple pus method using the image intensitiesand a clustering algorithm for the newest neighbors. This work was directed at segmentationof Multiple Schlerosis patients with extensive plaques and peri-ventricular effusions whoreceived multiple MR examinations over time.In this study. the fractal methodology at a local level was investigated. In particular. theblanket method and variation method were used to calculate the global fractal dimension onknown surfaces and the local dimensions on the s m e surfices as well as magnetic resonanceimages of a canine mandible. These methods were also used to study texture discriminationand segmentation, where the effects of blanket thickness. local matrix size and number ofpoints used for the least squares calculation of the local dimension were examined. It wasfound that the global fractal dimensions had an average agreement of 3.0 for known Takagisuficlcrs ranging from 1.1 to 1.9 for 256 rank matrices when using the blanket and variationmethods. On the other hand. it wm shown that for MR images. the variation method wasunacceptable for discrimination.....-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46677.pdf
Title and statement of responsibility and	:	ALGORITHMS ON MR IMAGING AND RF FIELD MAPS OF BIRDCAGE RESONATORS [Thesis]