Robust analysis of short echo time (1)H MRSI of human brain
1Department of Radiology, University of California San Francisco, San Francisco, CA 94121, USA.
Abstract:
Short echo time proton MR Spectroscopic Imaging (MRSI) suffers from low signal-to-noise ratio (SNR), limiting accuracy to estimate metabolite intensities. A method to coherently sum spectra in a region of interest of the human brain by appropriate peak alignment was developed to yield a mean spectrum with increased SNR. Furthermore, principal component (PC) spectra were calculated to estimate the variance of the mean spectrum. The mean or alternatively the first PC (PC(1)) spectrum from the same region can be used for quantitation of peak areas of metabolites in the human brain at increased SNR. Monte Carlo simulations showed that both mean and PC(1) spectra were more accurate in estimating regional metabolite concentrations than solutions that regress individual spectra against the tissue compositions of MRSI voxels. Back-to-back MRSI studies on 10 healthy volunteers showed that mean spectra markedly improved reliability of brain metabolite measurements, most notably for myo-inositol, as compared to regression methods.
Insights
This study introduces a novel method for enhancing Magnetic Resonance Spectroscopic Imaging (MRSI) data by coherently summing spectra. This technique improves signal-to-noise ratio (SNR) and metabolite quantification accuracy in the human brain.
Area of Science:
- Neuroimaging
- Biomedical Engineering
- Spectroscopy
Background:
- Short echo time proton MR Spectroscopic Imaging (MRSI) is limited by low signal-to-noise ratio (SNR), hindering accurate metabolite intensity estimation.
- Existing methods for metabolite quantification in MRSI can be less reliable due to inherent data noise.
Purpose of the Study:
- To develop and validate a method for improving SNR in human brain MRSI data.
- To enhance the accuracy of metabolite quantification using the improved MRSI data.
Main Methods:
- A novel technique involving coherent summation of spectra within a region of interest, with appropriate peak alignment, was developed.
- Principal Component (PC) spectra, including the first PC (PC(1)), were calculated to assess spectral variance.
- Monte Carlo simulations and back-to-back MRSI studies on 10 healthy volunteers were conducted for validation.
Main Results:
- The developed method yields a mean spectrum with significantly increased SNR compared to individual spectra.
- Both mean and PC(1) spectra demonstrated higher accuracy in estimating regional metabolite concentrations than regression-based methods.
- Mean spectra markedly improved the reliability of brain metabolite measurements, particularly for myo-inositol.
Conclusions:
- Coherent spectral summation is an effective strategy to boost SNR in short echo time proton MRSI.
- The enhanced SNR improves the reliability and accuracy of metabolite quantification in the human brain.
- This method offers a more robust approach for analyzing MRSI data compared to traditional regression techniques.
