Document Details

Document Type : Thesis 
Document Title :
MAGNETIC RESONANCE IMAGE RECONSTRUCTION FROM PARTIAL FOURIER DATA
بناء صورة الرنين المغناطيسي من بيانات فورير الجزئية
 
Subject : Faculty of Engineering 
Document Language : Arabic 
Abstract : Magnetic Resonance Imaging ( MRI ) outperforms other imaging modalities in contrast and resolution, specially for soft tissues. The time consumed for imaging is considerably long, which increases operational costs, limits MRI application and introduces artefacts that degrade image quality. Among the causes of image artefacts is motion of the imaged object, that is boosted by the long time of imaging. This is particularly true in areas where object or physiological motion is present, as in cardiac, thoracic, abdomen and pediatric imaging. The applications of MRI are limited by the lengthened imaging time in areas where imaging speed need to be fast enough to catch fast biological changes as in functional brain imaging, for example. This Thesis aims to optimize partial Fourier reconstruction (PFR) methods used in MRI as a mean for shortening imaging time. The researcher followed an inductive approach, starting with working on the data to get to a better optimised results. The focus was on Projection Onto Convex sets algorithm (POCs), particularly on filters implementation used for smoothing and merging Fourier data before the transformation to image domain takes place. A performance comparison, based on mean square error (MSE) between partially and fully reconstructed MRI images, was conducted for Eighteen symmetrical smoothing filters besides the shape of merging filter. In addition, a qualitative comparison between the original and enhanced POCS reconstructed MRI images was made, and showed a relative better performance for the latter. This Thesis concludes the feasibility of optimization for the current PFR existed in the literature. 
Supervisor : Dr.. Yasser Qadah 
Thesis Type : Master Thesis 
Publishing Year : 1442 AH
2020 AD 
Added Date : Friday, January 29, 2021 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
عبدالعزيز أبو الفضل الحيدريAl-Haidari, Abdulaziz Abu AlFadlResearcherMaster 

Files

File NameTypeDescription
 46868.pdf pdf 

Back To Researches Page