Local Asymmetric Least Squares (LAsLS) for Noisy and Complex Baseline Correction
Journal of Chemometrics · 2026-08 · 40 (8)
DOI: 10.1002/cem.70169 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemometrics, online erschienen am 29.07.2026. Verfasst von Adrián Gómez‐Sánchez und 2 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The asymmetric least squares (AsLS) method is commonly used and works well when baseline characteristics are relatively uniform, but can be less accurate for non‐uniform, complex, or noisy baselines. Here, we present local AsLS (LAsLS), a local extension of AsLS that preserves the original penalized least‐squares framework while allowing asymmetric weights and smoothing penalties to vary across user‐defined regions of the signal. This local parameterization provides more accurate baseline estimation for heterogeneous baselines. The method is versatile and can be applied to a variety of techniques, such as chromatography, Raman spectroscopy, and mid‐infrared spectroscopy. To test the method under controlled conditions, we compared LAsLS with traditional AsLS using simulated chromatogram‐like signals that mimic complex baseline scenarios. Across four cases with heterogeneous baselines, AsLS baseline rRMSE was 2.6–5.7 times higher than LAsLS, and peak quantification rRMSE was 1.4–3.5 times higher. In a controlled PLSR quantitation example, LAsLS also reduced external prediction rRMSE compared with AsLS (11.3% vs. 17.9%). We provide open‐source MATLAB implementations with a graphical interface to facilitate method adoption.
Zitieren
APA
Gómez‐Sánchez, A., Torres‐Cobos, B., & Oliveira, R. R. D. (2026). Local Asymmetric Least Squares (LAsLS) for Noisy and Complex Baseline Correction. Journal of Chemometrics, 40(8). https://doi.org/10.1002/cem.70169
Vancouver
Gómez‐Sánchez A, Torres‐Cobos B, Oliveira RRD. Local Asymmetric Least Squares (LAsLS) for Noisy and Complex Baseline Correction. Journal of Chemometrics. 2026;40(8). doi:10.1002/cem.70169
BibTeX
@article{GmezSnchez2026,
author = {Gómez‐Sánchez, Adrián and Torres‐Cobos, Berta and Oliveira, Rodrigo Rocha de},
title = {Local Asymmetric Least Squares (LAsLS) for Noisy and Complex Baseline Correction},
journal = {Journal of Chemometrics},
year = {2026},
volume = {40},
number = {8},
doi = {10.1002/cem.70169}
}
EUVIMED