A Fully Customized Baseline Removal Framework for Spectroscopic Applications

Abstract

The task of proper baseline or continuum removal is common to nearly all types of spectroscopy. Its goal is to remove any portion of a signal that is irrelevant to features of interest while preserving any predictive information. Despite the importance of baseline removal, median or guessed default parameters are commonly employed, often using commercially available software supplied with instruments. Several published baseline removal algorithms have been shown to be useful for particular spectroscopic applications but their generalizability is ambiguous. The new Custom Baseline Removal (Custom BLR) method presented here generalizes the problem of baseline removal by combining operations from previously proposed methods to synthesize new correction algorithms. It creates novel methods for each technique, application, and training set, discovering new algorithms that maximize the predictive accuracy of the resulting spectroscopic models. In most cases, these learned methods either match or improve on the performance of the best alternative. Examples of these advantages are shown for three different scenarios: quantification of components in near-infrared spectra of corn and laser-induced breakdown spectroscopy data of rocks, and classification/matching of minerals using Raman spectroscopy. Software to implement this optimization is available from the authors. By removing subjectivity from this commonly encountered task, Custom BLR is a significant step toward completely automatic and general baseline removal in spectroscopic and other applications.

Shifted Excitation Raman Difference Spectroscopy with Charge-Shifting Charge-Coupled Device (CCD) Lock-In Detection

“…Due to the potential severity of the fluorescence problem, a wide range of fluorescence rejection techniques has been developed which can be coarsely divided into computational and experimental approaches. 1 The mathematical methods 2,3 are usually relatively easy to implement and do not require setup modifications but have limited capability in fluorescence removal. Temporal gating techniques 4,5 can be very powerful as they exploit the fact that Raman scattering and fluorescence occur on different time scales.…”

Developing Fully Automated Quality Control Methods for Preprocessing Raman Spectra of Biomedical and Biological Samples

“…35 Not only can these algorithms be combined in a great variety of preprocessing sequences, but new methods to perform a specific preprocessing operation can also be synthesized by combining subtask-executing elements from extant algorithms. 44 Such a large variety of preprocessing approaches can lead to an equally inappropriate range of outcomes. Again, an assessment of outcome quality is imperative.…”

Comparison of univariate and multivariate models for prediction of major and minor elements from laser-induced breakdown spectra with and without masking

“…Depending on the trial, baseline removal is then applied. For this paper, we used custom baseline removal (Custom BLR) as described in Giguere et al [34] and Dyar et al [35]. The method generalizes the problem of baseline removal by combining operations from previously proposed methods to synthesize new correction algorithms for each application and training set.…”

“…The method generalizes the problem of baseline removal by combining operations from previously proposed methods to synthesize new correction algorithms for each application and training set. Custom BLR creates novel methods, discovering new algorithms that maximize the predictive accuracy of the resulting spectroscopic models and yield significant improvements over existing methods [34].…”

“…Finally, the team's approach uses sub-models, in which different models are trained for varying ranges of composition for each element, and then blended back together. Our analysis used custom baseline removal [34,35] rather than the team's algorithm and our data pre-processing sequence is slightly modified, but it uses all channels of all spectra available. This makes the data set we analyzed rather more heterogeneous than the one actually used by the team.…”