Baseline roll removal in NMR spectra using bayesian analysis

In this paper we demonstrate the use of Bayesian analysis methods for the removal of baseline roll from NMR spectra. Baseline roll occurs in NMR spectra due to experimental artifacts in the first few channels of the acquired time-domain signal. Bayesian analysis is used to predict improved estimates of these values which are substituted into the experimental time-domain signal prior to Fourier transformation. Although applied to NMR spectra, the method is general and may be applied to baseline correction in most spectroscopic techniques.