Meeting Banner
Abstract #2017

Accelerated Phosphorus MR Spectroscopic Imaging of Human Brain Using Compressed Sensing

Gokce H. Hatay1, Cagil Gumus1, Bahattin Hakyemez2, Esin Ozturk-Isik3

1Electrical and Electronics Engineering, Yeditepe University, Istanbul, Turkey; 2Department of Radiology, Uludag University, Bursa, Turkey; 3Biomedical Engineering, Yeditepe University, Istanbul, Turkey

This study aims at assessing compressed sensing reconstruction for faster phosphorus MR spectroscopic imaging (31P MRSI) of human brain. A 31P MRSI dataset containing a tumor region and a healthy region was created based on the metabolite peak intensities of a volunteer dataset acquired at 3T. k-space data was randomly undersampled while preserving the cental portion for a reduction factor of 1.6, and reconstructed using compressed sensing. This study showed that compressed sensing reconstruction could be applied for faster 31P MR spectroscopic imaging, and its performance will be assessed in brain tumors.

Keywords

accelerate accelerated acid acquiring acquisition acquisitions adequate amplitude amplitudes application applied array averaging bias biomedical bland body bottom brain bursa career central chosen clearly coil compressed consent considered containing correction creating curie dataset datasets degradation detect development diagnosed disk domain dwell eight electrical electronics empirically energetic engineering especially faster filter frequency future grant healthy human implemented important in vivo included informed inverse investigate larger length limitations lobe localization long looked major measure medical membrane metabolite noise norm original outliers package parallel part patients pattern peak peaks penalty performance phosphorus preserving processed proposed proton quantified radiology random reconstruct reconstructed reconstruction reduced regarding removal requires respectively rest resultant scanned scanner selected sensing sensitive separable several significantly similarity similarly simulated slightly software space spectra spectral spectroscopic spectroscopy spectrum statistical status studies successfully supported surface survey synthesis systems table tissue transform transformed transition tumor tumors turkey variation volunteer water wavelet