Abstract
PurposeLow-rank denoising of MRSI data results in an apparent increase in spectral SNR. However, it is not clear if this translates to a lower uncertainty in metabolite concentrations after spectroscopic fitting. Estimation of the true uncertainty after denoising is desirable for downstream analysis in spectroscopy. In this work, the uncertainty reduction from low-rank denoising methods based on spatiotemporal separability and linear predictability in MRSI are assessed. A new method for estimating metabolite concentration uncertainty after denoising is proposed. Automatic rank threshold selection methods are also assessed in simulated low SNR regimes.MethodsAssessment of denoising methods is conducted using Monte Carlo simulation of proton MRSI data and by reproducibility of repeated in vivo acquisitions in 5 subjects.ResultsIn simulated and in vivo data, spatiotemporal based denoising is shown to reduce the concentration uncertainty, but linear prediction denoising increases uncertainty. Uncertainty estimates provided by fitting algorithms after denoising consistently underestimate actual metabolite uncertainty. However, the proposed uncertainty estimation, based on an analytical expression for entry-wise variance after denoising, is more accurate. It is also shown automated rank threshold selection using Marchenko-Pastur distribution can bias the data in low SNR conditions. An alternative soft-thresholding function is proposed.ConclusionLow-rank denoising methods based on spatiotemporal separability do reduce uncertainty in MRS(I) data. However, thorough assessment is needed as assessment by SNR measured from residual baseline noise is insufficient given the presence of non-uniform variance. It is also important to select the right rank thresholding method in low SNR cases.
Highlights
Increasing the SNR of MRSI allows for faster acquisitions, more reliable quantification, or higher resolution acquisitions
Qualitative evaluation of the variance estimated by the proposed method against that measured by Monte Carlo (MC) simulation showed that the proposed method slightly underestimated the variance, but predicted its non-u niform time dependence, as well as capturing features of the covariance structure (Supporting Information Figures S6 and S7)
This work has demonstrated that low-rank denoising based on spatiotemporal separability can reduce uncertainty of estimated metabolite concentrations in both synthetic and in vivo 1H-M RSI data
Summary
Increasing the SNR of MRSI allows for faster acquisitions, more reliable quantification, or higher resolution acquisitions. Low-rank denoising methods achieve this in spectroscopic imaging data either by exploiting the linear predictability, or the spatiotemporal separability of the spectroscopic data, or both.[1] Low-rank denoising is a data-driven technique and does not incorporate prior knowledge or physical models of the data. These methods have recently been applied to MR imaging techniques that use an additional dimension of encoding, such diffusion encoding direction in diffusion-weighted MRI,[2,3] and time in functional MRI.[4] Low-rank models have been applied directly in the reconstruction of fast MRSI acquisitions.[5,6] The application of low-rank models to these techniques aims to exploit signal correlations across these encoding dimensions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
