Quantification of in vivo short-echo proton magnetic resonance spectroscopy of the human brain on a clinical whole-body 5-T platform under two different macromolecule modeling approaches.
- 1MR Application Development Department, Shanghai United-Imaging Healthcare Co., Ltd., Shanghai, China.
- 2Shanghai United-Imaging Healthcare Advanced Technology Research Institute Co., Ltd., Shanghai, China.
Abstract
Short-echo proton magnetic resonance spectroscopy (1H MRS) offers valuable insights into endogenous brain metabolites, commonly referred to as the neurochemical profile, enabling identification of critical biomarkers for neurological health and disease. However, the reliable extraction of these quantitative neurochemical profiles at high magnetic fields is significantly challenged by the presence of the underlying macromolecular (MM) background signal. As field strength increases to 5 T, the MM components appear with greater resolution and complexity. Although the effects of MM signals on neurochemical profiling at 3 and 7 T are well documented, their corresponding performance related to neurochemical profiling at 5 T remains uncharacterized. Compared to 3- and 7-T platforms, 5-T platforms theoretically offer a unique balance between signal-to-noise ratio (SNR) and field homogeneity. This study aimed to evaluate the quality of 5-T short-echo proton magnetic resonance (1H MR) spectra and to compare the effects between incorporating in vivo MMs and simulated MMs on neurochemical profiling. Short-echo 1H MR spectra [echo time (TE) =10 ms] were acquired on healthy volunteers at 5-T with a standard stimulated echo acquisition mode (STEAM) sequence. An in vivo MM baseline was first measured via an optimized inversion recovery STEAM (IR-STEAM) sequence [inversion time (TI) =750 ms] and postprocessed to remove residual metabolite signals. Spectral fitting was performed with a linear combination model (LCModel) under two strategies: (I) the experimentally derived in vivo MM baseline and (II) default simulated Voigt functions. Quantification reliability was assessed according to Cramér-Rao lower bounds (CRLBs) and intermethod concentration differences. The 5-T platform yielded high-quality MM spectra for the in vivo MM baseline, along with spectra with a mean SNR of ~110 and a metabolic linewidth of ~6 Hz for the quantification of the neurochemical profile. Both MM approaches successfully resolved 16 metabolites. Although the default simulated MM functions in the LCModel provided acceptable approximations for high-abundance metabolites [e.g., glutamate (Glu), N-acetylaspartate (NAA), and total creatine (tCr)]-12 neurochemicals with a mean ≤20%-they failed to replicate subtle spectral features at 2.5, 3.2, and 3.5-3.9 ppm. The in vivo MM approach significantly improved the quantification of low-concentration metabolites [e.g., glutamine (Gln), taurine (Tau), ascorbate (Asc), and scyllo-inositol (Scyllo)] and enhanced fitting residual performance. Through use of the in vivo MM baseline, 14 neurochemicals were reliably quantified, with a mean CRLB ≤20%. This study validated the 5-T platform as a high-performance environment for neurochemical profiling, offering an optimal intermediate between 3 and 7 T. Although simulated MM models are sufficient for major metabolites, a refined in vivo MM baseline is necessary for comprehensive and accurate quantification of the neurochemical profile. This framework provides a robust methodological foundation for future clinical research into metabolic alterations in brain disorders at 5 T.
Related
- Observer-Related Versus Cross-Modality Variability in Ventricular Quantification by Cardiac CT and Cardiovascular Magnetic Resonance
- Esophageal resonance Raman spectroscopy tracks mixed venous oxygenation in experimental shock and hypoxemia
- Temporal Magnetic Resonance Imaging Changes in Acute Hypoglycemic Encephalopathy: Two Cases With Different Gray Matter Lesion Patterns.
- Dual-Energy Computed Tomography Versus Conventional Magnetic Resonance In-Phase/Opposed-Phase Imaging for Liver Fat Fraction Quantification: A Retrospective Cohort Study.
- Relic of the Past or Clinical Compass? Re-Evaluating the Diagnostic Utility of Proton Magnetic Resonance Spectroscopy (1H-MRS) in Encephalitis and Infectious CNS Diseases
BibTeX
@article{wen2026quantification,
title = {{Quantification of in vivo short-echo proton magnetic resonance spectroscopy of the human brain on a clinical whole-body 5-T platform under two different macromolecule modeling approaches.}},
author = {Linfei Wen and Bin Deng and Aiguo Xue and Hongxia Lei},
journal = {Quantitative imaging in medicine and surgery},
year = {2026},
month = sep,
doi = {10.21037/qims-2025-1-2810},
url = {https://europepmc.org/article/MED/42997954},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Quantification of in vivo short-echo proton magnetic resonance spectroscopy of the human brain on a clinical whole-body 5-T platform under two different macromolecule modeling approaches.
AU - Linfei Wen
AU - Bin Deng
AU - Aiguo Xue
AU - Hongxia Lei
PY - 2026
DA - 2026/09/29
JO - Quantitative imaging in medicine and surgery
DO - 10.21037/qims-2025-1-2810
UR - https://europepmc.org/article/MED/42997954
AB - Short-echo proton magnetic resonance spectroscopy (1H MRS) offers valuable insights into endogenous brain metabolites, commonly referred to as the neurochemical profile, enabling identification of critical biomarkers for neurological health and disease. However, the reliable extraction of these quantitative neurochemical profiles at high magnetic fields is significantly challenged by the presence of the underlying macromolecular (MM) background signal. As field strength increases to 5 T, the MM components appear with greater resolution and complexity. Although the effects of MM signals on neurochemical profiling at 3 and 7 T are well documented, their corresponding performance related to neurochemical profiling at 5 T remains uncharacterized. Compared to 3- and 7-T platforms, 5-T platforms theoretically offer a unique balance between signal-to-noise ratio (SNR) and field homogeneity. This study aimed to evaluate the quality of 5-T short-echo proton magnetic resonance (1H MR) spectra and to compare the effects between incorporating in vivo MMs and simulated MMs on neurochemical profiling. Short-echo 1H MR spectra [echo time (TE) =10 ms] were acquired on healthy volunteers at 5-T with a standard stimulated echo acquisition mode (STEAM) sequence. An in vivo MM baseline was first measured via an optimized inversion recovery STEAM (IR-STEAM) sequence [inversion time (TI) =750 ms] and postprocessed to remove residual metabolite signals. Spectral fitting was performed with a linear combination model (LCModel) under two strategies: (I) the experimentally derived in vivo MM baseline and (II) default simulated Voigt functions. Quantification reliability was assessed according to Cramér-Rao lower bounds (CRLBs) and intermethod concentration differences. The 5-T platform yielded high-quality MM spectra for the in vivo MM baseline, along with spectra with a mean SNR of ~110 and a metabolic linewidth of ~6 Hz for the quantification of the neurochemical profile. Both MM approaches successfully resolved 16 metabolites. Although the default simulated MM functions in the LCModel provided acceptable approximations for high-abundance metabolites [e.g., glutamate (Glu), N-acetylaspartate (NAA), and total creatine (tCr)]-12 neurochemicals with a mean ≤20%-they failed to replicate subtle spectral features at 2.5, 3.2, and 3.5-3.9 ppm. The in vivo MM approach significantly improved the quantification of low-concentration metabolites [e.g., glutamine (Gln), taurine (Tau), ascorbate (Asc), and scyllo-inositol (Scyllo)] and enhanced fitting residual performance. Through use of the in vivo MM baseline, 14 neurochemicals were reliably quantified, with a mean CRLB ≤20%. This study validated the 5-T platform as a high-performance environment for neurochemical profiling, offering an optimal intermediate between 3 and 7 T. Although simulated MM models are sufficient for major metabolites, a refined in vivo MM baseline is necessary for comprehensive and accurate quantification of the neurochemical profile. This framework provides a robust methodological foundation for future clinical research into metabolic alterations in brain disorders at 5 T.
SN - 2223-4292
SN - 2223-4306
ER -