Meeting Banner
Abstract #4464

Proton Magnetic Resonance Spectroscopic Imaging (MRSI) in a Clinical Setting: A Scheme to Reduce Chemical Shift VOI Misregistration-Induced Metabolite Ratio Bias in Edge Voxels

David J. Manton1, Lawrence Kenning1, Ralph Noeske2, Timo Schirmer3, Gary P. Liney4, Martin Lowry1, Martin D. Pickles1, Richard Bartlett5, Christopher Rowland-Hill

1YCR Centre for MR Investigations, The University of Hull, Hull, East Yorkshire, United Kingdom; 2Applied Science Laboratory Europe, GE Healthcare, Berlin, Germany; 3Applied Science Laboratory Europe, GE Healthcare, Muenchen, Germany; 4Radiation Physics, Hull & East Yorkshire Hospitals NHS Trust, Cottingham, East Yorkshire, United Kingdom; 5Radiology Department, Hull & East Yorkshire Hospitals NHS Trust, Hull, East Yorkshire, United Kingdom

Clinical PRESS-based proton magnetic resonance spectroscopic imaging (MRSI) choline:NAA ratios demonstrated chemical shift volume of interest misregistration-induced bias from 42% too low to 66% too high over a 79 cm 2D VOI in a homogenous phantom at 3 T. A phantom-based calibration protocol was developed to allow robust in vivo bias correction where other ameliorating methods are unavailable (e.g. on clinical scanners) or impractical (e.g. with tumours lying close to the skull). Water resonant frequency changed by 25 Hz between in vivo (body temperature) and phantom (room temperature) experiments and this was taken into account when calculating the calibration map.

Abstract PDF E-Poster Video

Keywords

address adiabatic adjacent adjustment allowed allowing ameliorated anatomy application applied appreciable aqueous arising available back bands bandwidth bandwidths bias body brain buffer calibration cancer carried central chemical choline clinical column columns contamination contours corrected correction default defining dependent developed divided dummy east edge enlarged ensure erroneous errors example excitation experiment facilities failure faithful field filter final frequency generated geometry give gives grid healthy hospitals hull in vivo induced investigations kenning kind kingdom known laboratory lead leading lipid martin measured measuring median metabolite metabolites model need noted object offset optimal outcome paired phantom physics pickles planning potentially presence press previously prohibited protocol proton pulse pulses radiation radiology radiotherapy reduce representation rise robust room rows saturation scanner scheme science select setting sizing skull slice smoothing solution spatial spectra spectral spectroscopic spherical statistically studies systematic systems target tech temperature threshold treat treatment trimmed trust unwelcome upon utility view vitro volume volumes volunteer water whereby width zone