New Hybrid Algorithm for Maintaining Multivariate Quantitative Calibrations of a Near-Infrared Spectrometer

Abstract

Our newly developed prediction-augmented classical least-squares/partial least-squares (PACLS/PLS) hybrid algorithm can correct for the presence of unmodeled sources of spectral variation such as instrument drift by explicitly incorporating known or empirically derived information about the unmodeled spectral variation. We have tested the ability of the new hybrid algorithm to maintain a multivariate calibration in the presence of instrument drift using a near-infrared (NIR) spectrometer (7500–11 000 cm−1) to quantitate dilute aqueous solutions containing glucose, ethanol, and urea. The spectral variations required to update the multivariate models for both short- and long-term drift were obtained using a single representative midpoint sample whose spectrum was repeatedly measured during collection of calibration data and during collection of separate validation sample spectra on three subsequent days. The performance of the PACLS/PLS model for maintaining a calibration was compared to PLS with subset recalibration, a method that has previously been applied to maintenance and transfer of calibration. Without drift corrections, both PACLS/PLS and PLS had poor predictive ability on sample spectra collected on subsequent days. Unlike previous maintenance of calibration studies that corrected for long-term drift only, the PACLS/PLS and PLS models demonstrated the best predictive abilities when short-term drift was also corrected. The PACLS/PLS hybrid model outperformed PLS with subset recalibration for near real-time predictions when instrument drift was determined from the repeat samples closest in time to the measurement of the unknown. Near real-time standard errors of prediction (SEPs) for the hybrid model were comparable to the cross-validated SEPs obtained with the original calibration model.

Impact of standardization sample design on Tikhonov regularization variants for spectroscopic calibration maintenance and transfer

“…In this case, y L ¼ 0. Similarly, key eigenvectors from the singular value decomposition (SVD) of L with samples containing constant or no analyte, such as spectra from repeatedly measuring the spectrum of a single sample [15,16], could be used for L with y L ¼ 0.…”

Exploration of new multivariate spectral calibration algorithms.

“…Since the generalized ACLS algorithm is compatible with the PACLS method that allows rapid updating of models during prediction, the powerful combination of PACLS with ACLS is very promising for rapidly maintaining and transferring models for system drift, spectrometer differences, and unmodeled components without the need for recalibration. [4,5,161 The comparisons under different noise assumptions in the simulations obtained during this investigation emphasize the need to use realistic simulations when making comparisons between various multivariate calibration methods. Clearly, the conclusions of the relative performance of various methods were found to be dependent on how realistic the spectral errors were in the simulated data.…”

“…[I] In results published from this current LDRD project, we have demonstrated that the new hybrid algorithm can be used both to improve prediction abilities when instrument drift is present [2] and when transferring calibrations between spectrometers [3]. We demonstrated that the new hybrid method is very successful at maintaining and transferring multivariate spectral calibrations and that they performed as well or better than standard partial least squares (PLS) multivariate calibrations when calibration maintenance andlor transfer were required.…”

“…We have called this new class of multivariate calibration methods augmented classical least squares (ACLS) calibrations, and we have recently published the theory behind these algorithms and given results with experimental and simulated data. [4,51 (See also a review of these methods by Malinowski [6]). These new methods extend the traditional classical least squares (CLS) method to handle data sets where all spectrally active analytes are not explicitly included in the calibration.…”

“…When used in conjunction with prediction augmented CLS (PACLS), [7] the ACLS calibration models can be efficiently and quickly updated to handle maintenance and transfer of calibration issues without recalibration. [4,51 In this report, we will focus primarily on those studies that have not yet been published. Thus, we will present the theory behind the new family of ACLS algorithms, discuss new options for factor selection for these methods, apply ACLS to experiments requiring transfer of ACLS calibration models between different spectrometer configurations, and compare the relative merits of the ACLS with the more commonly used PLS models using a variety of simulated data sets.…”