English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Poster

The necessity of parametrizing macromolecules for accurate quantification of ultra-short TE and TR 1H FID MRSI data at 9.4T

MPS-Authors
/persons/resource/persons192740

Nassirpour,  S
Max Planck Institute for Biological Cybernetics, Max Planck Society;
Research Group MR Spectroscopy and Ultra-High Field Methodology, Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons192839

Chang,  P
Research Group MR Spectroscopy and Ultra-High Field Methodology, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons84402

Henning,  A
Max Planck Institute for Biological Cybernetics, Max Planck Society;
Research Group MR Spectroscopy and Ultra-High Field Methodology, Max Planck Institute for Biological Cybernetics, Max Planck Society;

External Resource

Link
(Publisher version)

Fulltext (public)
There are no public fulltexts stored in PuRe
Supplementary Material (public)
There is no public supplementary material available
Citation

Nassirpour, S., Chang, P., & Henning, A. (2017). The necessity of parametrizing macromolecules for accurate quantification of ultra-short TE and TR 1H FID MRSI data at 9.4T. Poster presented at 34th Annual Scientific Meeting of the European Society for Magnetic Resonance in Medicine and Biology (ESMRMB 2017), Barcelona, Spain.


Cite as: http://hdl.handle.net/21.11116/0000-0000-C3F7-D
Abstract
Purpose/Introduction: At ultra-short TE and TR, macromolecule quantification becomes extremely important [1, 2], because of shorter relaxation times and thus lower saturation, these molecules contribute greatly to the spectrum. Measuring the MMs and including them as a single baseline might not be enough because of regional and relaxation time differences between the different macromolecule components [1, 2]. In this work, we measured and modelled individual macromolecule components at 9.4T. We then used our proposed MM model to highlight the necessity of parametrizing the individual macromolecule components for ultra-short TE and TR MRSI measurements. Subjects and Methods: 8 healthy subjects were scanned on a 9.4T Siemens whole-body human scanner. Metabolite-nulled data using DIR [1] were acquired and used to model the MM baseline. The T1 of these metabolites in gray and white matter were estimated using variable flip angle method on MRSI datasets acquired at 4 different flip angles with a TR of 500 ms. High resolution (3.25mm 9 3.25mm 9 10 mm) ultra-short TE 1H FID MRSI [3–6] data with two different TRs of 300 ms and 500 ms were also collected. These data were fitted using LCMODEL [7] with three different MM baseline models (‘‘no MM’’, ‘‘measured MM’’ as an average single-component MM baseline model, and our proposed multi-component ‘‘modelled MM’’ shown in Figure 1a). Results: Figure 1 shows our proposed MM model along with the estimated T1 values. Figure 2 shows the metabolite maps resulting from fitting the datasets acquired at TR = 300 ms with 3 different models for 2 different volunteers. Figure 3 shows a boxplot of how much the metabolite concentrations are overestimated when different models are used. The difference between the two different TRs are due to saturation effects that have not been corrected for. However, the figure highlights how using different MM models on each TR can account for the macromolecule saturation effects. Discussion/Conclusion: Figure 1 shows that there are regional and relaxation time differences between individual macromolecule components that cause metabolite concentrations to differ from the no MM model to various degrees (as shown in Figure 2). Figure 3 further shows that if a single MM baseline is used in fitting, depending on the TR, the concentration of metabolite will be underestimated to varying degrees, since different macromolecule components relax at different rates. Also, due to regional differences, the overestimation degree varies between white and gray matter. The only way to account for all of this is to use a parametrized MM model like the one we have proposed.