Factors affecting the quantification of short echo in-vivo H-1 MR spectra:prior knowledge, peak elimination, and filtering

Short echo H-1 in-vivo brain MR spectra are difficult to quantify for sever al reasons: low signal to noise ratio, the severe overlap of spectral lines , the presence of macromolecule resonances beneath the resonances of intere st, and the effect of resonances adjacent to the spectral region of interes t (SRI). This paper outlines several different quantification strategies an d the effect of each on the precision of in-vivo metabolite measurements. I n-vivo spectra were quantified with no operator interaction using a templat e of prior knowledge determined by mathematically modeling separate in-vitr o metabolite spectra. Metabolite level estimates and associated precision w ere compared before and after the inclusion of macromolecule resonances as part of the prior knowledge, and following two different methods of handlin g resonances adjacent to the SRI. The effects of rectangular and exponentia l filters were also investigated. All methods were tested using repeated in -vivo spectra from one individual acquired at 1.5 T using stimulated echo a cquisition mode (STEAM, TE = 20 ms) localization. The results showed that t he inclusion of macromolecules in the prior knowledge was necessary to obta in metabolite levels consistent with the literature, while the fitting of r esonances adjacent to the SRI concurrent with modeled metabolites optimized the precision of metabolite estimates. Metabolite levels and precision wer e also affected by rectangular and exponential filtering, suggesting cautio n must be taken when such filters are used. Copyright (C) 1999 John Wiley & Sons, Ltd.