Files in this item



application/pdfECE499-Sp2019-park.pdf (4MB)Restricted to U of Illinois
(no description provided)PDF


Title:Subspace approach to magnetic resonance imaging
Author(s):Park, Somie
Contributor(s):Liang, Zhi-Pei
partial separability
subspace approach
Abstract:Magnetic resonance spectroscopic imaging (MRSI) is an imaging method that uses the same principles as MRI and detects signals from water, lipids, brain metabolites and neurotransmitters. However, unlike MRI, it gives a time sequence of images so that each voxel gives a frequency spectrum where certain peaks correspond to different metabolites. This gives more information than MRI and has applications in cancer imaging and detection/characterization of disease. However, long data acquisition time, poor spatial resolution, and low SNR have hindered clinical usage of MRSI. The subspace approach was developed by Professor Liang’s research group that combats these challenges. The subspace approach takes advantage of that the desired spatial-temporal function of the metabolite signal can be modeled by partially separable functions. With regards to data acquisition, the subspace approach has sparse temporal sampling and extended k-space coverage which allows for accelerated data acquisition. The subspace approach reduces the number of unknowns and enables the accurate reconstruction of high resolution spatiospectral function from sparse and noisy data. Using this approach results in faster scan times while still retaining the useful metabolic information.
Issue Date:2019-05
Date Available in IDEALS:2019-06-17

This item appears in the following Collection(s)

Item Statistics