MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

QC of Patient MR Spectroscopy Can Identify Sub-Optimal Data and System Malfunctions

SA Einstein*, P Hou , SJ Fahrenholtz , RJ Stafford , UT MD Anderson Cancer Center, Houston, TX

Presentations

(Sunday, 7/29/2018) 4:00 PM - 4:55 PM

Room: Room 207

Purpose: MR spectroscopy (MRS) is a clinical technique that is rapidly expanding in use for diagnosing a variety of conditions. There is currently a dearth of information published on the quality of MRS in patients on clinical systems. Current MRS QC recommended by AAPM TG reports are phantom-based, which is not ideal and cannot identify poor patient data. This study investigated the feasibility of using no-cost fitting software to perform QC on patient MRS.

Methods: Patient brain lesion and contralateral MRS data (single voxel) were acquired from a Discovery MR750 (n=16; GE Healthcare, Chicago, IL), Discovery MR750w (n=8, GE Healthcare), and MAGNETOM Prisma (n=16; Siemens Healthineers, Erlangen, Germany) using a PRESS sequence with similar acquisition parameters. Spectra were processed with TARQUIN (http://tarquin.sourceforge.net/) using the default basis sets and parameters. No eddy current correction was performed. Parameters of interest included the standard deviation of the fit residual divided by the standard deviation of the noise (Q), full-width at half maximum (FWHM) and signal-to-noise ratio (SNR) of the largest metabolite, and deviation of the fitted baseline from a straight line (BL). Significant differences were determined using Welch’s t-test.

Results: Mean values and standard deviations (SD) were successfully measured for all parameters on all systems. Inter-patient variability typically dominated over inter-scanner variability. In lesion spectra, the Prisma had significantly better Q than the MR750 and MR750w (p<0.005). The Prisma also had better SNR than the MR750w in both lesion and contralateral spectra (p<0.05). Several outliers (parameter value>2SD from mean) were identified and analyzed. Common causes for inferior performance were poor water suppression, poor shim, and crusher failure.

Conclusion: The output of MRS fitting software successfully quantified patient spectral quality and identified spectral artifacts. This technique is currently being transitioned to the clinic for real-time patient MRS QC and longitudinal monitoring of scanner performance.

Keywords

Not Applicable / None Entered.

Taxonomy

IM- MRI : Spectroscopy

Contact Email