Automated Principal Component-Based Orthogonal Signal Correction Applied to Fused Near Infrared���Mid-Infrared Spectra of French Olive Oils - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Analytical Chemistry Année : 2009

Automated Principal Component-Based Orthogonal Signal Correction Applied to Fused Near Infrared���Mid-Infrared Spectra of French Olive Oils

Résumé

An approach for automating the determination of the number of components in orthogonal signal correction (OSC) has been devised. In addition, a novel principal component OSC (PC-OSC) is reported that builds softer models for removing background from signals and is much faster than the partial least-squares (PLS) based OSC algorithm. These signal correction methods were evaluated by classifying fused near-and mid-infrared spectra of French olive oils by geographic origin. Two classification methods, partial least-squares-discriminant analysis (PLS-DA) and a fuzzy rule-building expert system (FuRES), were used to evaluate the signal correction of the fused vibrational spectra from the olive oils. The number of components was determined by using boot-strap Latin partitions (BLPs) in the signal correction routine and maximizing the average projected difference resolution (PDR). The same approach was used to select the number of latent variables in the PLS-DA evaluation and perfect classification was obtained. Biased PLS-DA models were also evaluated that optimized the number of latent variables to yield the minimum prediction error. Fuzzy or soft classification systems benefit from background removal. The FuRES prediction results did not differ significantly from the results that were obtained using either the unbiased or biased PLS-DA methods, but was an order of magnitude faster in the evaluations when a sufficient number of PC-OSC components were selected. The importance of bootstrapping was demonstrated for the automated OSC and PC-OSC methods. In addition, the PLS-DA algorithms were also automated using BLPs and proved effective. Measurement data are frequently plagued with background variations especially for large scale studies and for complex samples. Accurate background correction is a difficult yet important problem for data analysis. There are many approaches to correcting backgrounds or removing baseline variations. The key idea is to remove unwanted or irrelevant variances from the measurement data. Because background correction is a key first step for data processing, improper corrections can generate errors that propagate through the modeling process and result in inaccurate predictions. In many cases, background correction, such as polynomial curve-fitting, the practice is more a time-consuming art than a science. Adjustments to parameters such as polynomial order and window size can have a pronounced effect on the corrected data. Furthermore, spurious artifacts can be introduced that may cause problems later during model building and interpretation steps. With the modern trend of making analytical measurements of more samples that are complex, automated methods of model building and evaluation are essential. Embedding automated chemometric methods so that they are transparent to the user into analytical instrumentation is a key step toward the design of an intelligent chemical instrument.
Fichier non déposé

Dates et versions

hal-01493433 , version 1 (21-03-2017)

Identifiants

Citer

Peter de B Harrington, Jacky Kister, Jacques Artaud, Nathalie Dupuy. Automated Principal Component-Based Orthogonal Signal Correction Applied to Fused Near Infrared���Mid-Infrared Spectra of French Olive Oils. Analytical Chemistry, 2009, 81, pp.7160 - 7169. ⟨10.1021/ac900538n⟩. ⟨hal-01493433⟩
61 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More