Smoothness of in vivo spectral baseline determined by mean-square error

Abstract

Purpose A nonparametric smooth line is usually added to spectral model to account for background signals in vivo magnetic resonance spectroscopy (MRS). The assumed smoothness of the baseline significantly influences quantitative spectral fitting. In this paper, a method is proposed to minimize baseline influences on estimated spectral parameters. Methods In this paper, the non-parametric baseline function with a given smoothness was treated as a function of spectral parameters. Its uncertainty was measured by root-mean-squared error (RMSE). The proposed method was demonstrated with a simulated spectrum and in vivo spectra of both short echo time (TE) and averaged echo times. The estimated in vivo baselines were compared with the metabolite-nulled spectra, and the LCModel-estimated baselines. The accuracies of estimated baseline and metabolite concentrations were further verified by cross-validation. Results An optimal smoothness condition was found that led to the minimal baseline RMSE. In this condition, the best fit was balanced against minimal baseline influences on metabolite concentration estimates. Conclusion Baseline RMSE can be used to indicate estimated baseline uncertainties and serve as the criterion for determining the baseline smoothness of in vivo MRS.

Regional and tissue-specific differences in brain glutamate concentration measured by in vivo single voxel MRS

“…Conversely, an over-smoothed baseline often results in poor fitting, characterized by large fit residuals, because background signals are not well represented. In this study, the spectral baselines were determined by a recently proposed method (Zhang and Shen, in press). This method minimizes the error of both the fit residuals and the uncertainties arising from the interaction between the baseline and metabolite peaks.…”

“…This method minimizes the error of both the fit residuals and the uncertainties arising from the interaction between the baseline and metabolite peaks. In Table 2, the mean coefficients of variation were derived from Cramer-Rao Lower Bounds (CRLB) (Cavassila et al, 2001; Zhang and Shen, in press). Note that the MPFC spectra have larger CRLBs than the OCC spectra, echoing that the former have poorer spectral quality than the latter (see Figure 2).…”

Adaptive Baseline Fitting for1H MR Spectroscopy Analysis

“…In ABfit, these non-linear parameters are estimated using a simplified initial fit (step 2) and subsequently refined (step 4). More recently, Zhang and Shen showed that a measure of the baseline uncertainty is also a useful criterion to determine baseline smoothness for simulated and experimentally acquired data [41].…”

Fast computation of full density matrix of multispin systems for spatially localizedin vivomagnetic resonance spectroscopy

“…This is important for short TE spectral fitting [22][23][24][25] because there are always couplings between the estimated lineshape and the baseline. 18 Constraints based on the correct T 2 s can mitigate estimation uncertainties originated from the above-mentioned interactions.…”