A multi-stack framework in magnetic resonance imaging

Image processing;Inverse problems;Super-resolution;Image reconstruction;MRI

Shilling, Richard Zethward ; Electrical and Computer Engineering

University:Georgia Institute of Technology

Department:Electrical and Computer Engineering

关键词: Image processing;    Inverse problems;    Super-resolution;    Image reconstruction;    MRI;   

Others  :  https://smartech.gatech.edu/bitstream/1853/33807/1/shilling_richard_z_200905_phd.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Magnetic resonance imaging (MRI) is the preferred imaging modality for visualizationof intracranial soft tissues. Surgical planning, and increasingly surgical navigation, usehigh resolution 3-D patient-specific structural maps of the brain. However, the process ofMRI is a multi-parameter tomographic technique where high resolution imagery competesagainst high contrast and reasonable acquisition times.Resolution enhancement techniques based on super-resolution are particularly well suitedin solving the problems of resolution when high contrast with reasonable times forMRI acquisitions are needed. Super-resolution is the concept of reconstructing a high resolutionimage from a set of low-resolution images taken at dierent viewpoints or foci. TheMRI encoding techniques that produce high resolution imagery are often sub-optimal forthe desired contrast needed for visualization of some structures in the brain.A novel super-resolution reconstruction framework for MRI is proposed in this thesis.Its purpose is to produce images of both high resolution and high contrast desirable forimage-guided minimally invasive brain surgery. The input data are multiple 2-D multi-sliceInversion Recovery MRI scans acquired at orientations with regular angular spacing rotatedaround a common axis. Inspired by the computed tomography domain, the reconstruction isa 3-D volume of isotropic high resolution, where the inversion process resembles a projectionreconstruction problem. Iterative algorithms for reconstruction are based on the projectiononto convex sets formalism. Results demonstrate resolution enhancement in simulatedphantom studies, and in ex- and in-vivo human brain scans, carried out on clinical scanners.In addition, a novel motion correction method is applied to volume registration using aniterative technique in which super-resolution reconstruction is estimated in a given iterationfollowing motion correction in the preceding iteration. A comparison study of our methodwith previously published methods in super-resolution shows favorable characteristics of theproposed approach.