Technical reportManually Adjusted Versus Vendor-Preset Definition of Metabolite Boundaries: Impact on Proton Metabolite Ratios
Section snippets
Materials and methods
A total of 42 spectra spanning a range of normal and abnormal spectroscopic features and qualities were included in the study. The 42 spectra were derived from seven subjects, including three healthy adults with unremarkable brain MR examinations (one man and two women, age range 48−52 years, mean age 51 years) and four patients with a brain tumor (four men, age range 7−53 years, mean age 34 years). Institutional board review approval was obtained for the study.
Spectra were acquired using
Results
The preset peak boundaries rendered by the software in rank (i.e., frequency index number) and corresponding ppm were as follows: Cho: 67−77 (3.29−3.17 ppm), Cr: 76−89 (3.18−2.99 ppm), and NAA: 142−155 (2.20−2.00 ppm). These default preset peak boundaries were consistent in all subjects. The default central maxima for the three metabolites were Cho: 70 (3.27 ppm), Cr: 82 (3.09 ppm), and NAA: 145 (2.11 ppm). The manually adjusted peak boundaries in rank and corresponding ppm were as follows:
Discussion
As in vivo proton MR spectroscopy becomes a widely used tool in the characterization of focal brain lesions, assessing postprocessing methods and identifying potential pitfalls assumes increasing importance.
Our findings show that there are significant differences between metabolite ratios rendered by default software-rendered metabolite peaks and those estimated after manual adjustment of the peak boundaries by an experienced spectroscopist. Furthermore, the manually adjusted Cho/Cr and Cho/NAA
Conclusion
Our results suggest that there can be significant differences in metabolite ratios calculated using default and manually adjusted peak limits in proton MR spectroscopy. Furthermore, Cho/Cr and NAA/Cho adjusted metabolite ratios are closer to curve-fit values, which are considered the most accurate of the three.
References (8)
Reproducibility of metabolite peak areas in 1H MRS of brain
Magn Reson Imaging
(1996)- et al.
Improved method for accurate and efficient quantification of MRS data with use of prior knowledge
J Magn Reson
(1997) - et al.
Serial evaluation of patients with brain tumors using volume MRI and 3D 1H MRSI
NMR Biomed
(1999) In vivo molecular imaging for planning radiation therapy of gliomas: An application of 1H MRSI
J Magn Reson Imaging
(2002)
Cited by (4)
MR Spectroscopy Using Normalized and Non-normalized Metabolite Ratios for Differentiating Recurrent Brain Tumor from Radiation Injury
2011, Academic RadiologyCitation Excerpt :We determined the areas under the various metabolite peaks and calculated the ratios. The integration limits of the respective peaks were manually defined by the same neuroradiologist before computerized calculations (23). Multiple voxels covering the contrast-enhancing lesion were analyzed and the voxel with the most abnormal spectra in terms of changes in the metabolic peaks (ie, significant increase in Cho and/or significant decrease in NAA, Cho, and Cr peaks) were chosen as “diagnostic voxel” for the contrast-enhanced lesion.
Proton MR spectroscopy and white matter hyperintensities in idiopathic normal pressure hydrocephalus and other dementias
2010, British Journal of Radiology