• Progress in Medical Physics
• /
• v.11 no.1
• /
• pp.73-83
• /
• 2000

In this study, we investigate the effects of poor shimming on quantitative measurement of ratios of metabolite levels by proton magnetic resonance spectroscopy ($^1$H MRS). Coefficient of variation (COV) of metabolite ratios for point resolved spectroscopy (PRESS) and stimulated-echo acquisition mode (STEAM) spectra was evaluated from a phantom containing in vivo levels of metabolites using a conventional whole body 1.5T MR system and conventional acquisition and analysis protocol. A statistical P-value was also calculated from a linear regression for relationship of metabolite ratios. N-acetylaspartate (NAA)/ creatine (Cr) and NAA/ choline (Cho) had low COV values for the long and short TE spectra (29.1 and 27.5%; 23.8 and 12.6 %), whereas Cho/Cr and Cr/Cho had high COV values (50.0 and 68.6 %; 27.5 and 29.3 %). A linear relationship between NAA/Cr and Cho/Cr, and between NAA/Cho and Cr/Cho revealed the statistical significance in the long and short TE spectra, respectively (P < 0.0001 and P < 0.0001; P = 0.015 and P = 0.005). There was no significant relationship between Cho/NAA and Cr/NAA in the measurement (P = 0.159; P = 0.910). The present study suggested that NAA/Cr and NAA/Cho could be useful for data with poor shimming in $^1$H MR spectroscopy. In conclusion, statistical significance of metabolite ratios indicated that the Cr and Cho levels could be interpreted as a significant alteration factor in the long and short TE spectra, and then should be used with care to provide precise metabolite quantification.