Spectral Multivariate Calibration without Laboratory Prepared or Determined Reference Analyte Values

Local Modeling by Adapting Source Calibration Models to Analyte Shifted Target Domain Samples Without Reference Values

Spectral multivariate calibration aims to derive models characterizing mathematical relationships between sample analyte amounts and corresponding spectral responses. These models are effective at predicting target domain sample analyte amounts when target samples are within the analyte and spectral calibration source domain. Models fail when target samples shift (analyte amounts and/or spectra) from the original calibration domain model. A total recalibration solution requires acquisition of new sample reference values and spectra. However, obtaining enough reference values to distinguish the target domain may be challenging or expensive. A simpler approach adapts the original model to the target domain using target sample spectra without analyte reference values (unlabeled). Analytical chemists have developed several machine learning algorithms using unlabeled regression domain adaptation processes. Unfortunately, prediction accuracy declines for these methods depending on how much the target domain analyte distribution has shifted from the calibration distribution, and regression transfer learning methods are instead needed. Regression domain adaptation and transfer learning are often referred to as model updating in analytical chemistry, but regression domain adaptation only applies to spectral shifts. The regression transfer learning method presented in this paper named null augmentation regression constant analyte (NARCA) leverages unlabeled repeat spectra of a single target sample to update an original calibration model to the shifted target domain sample. With sample repeat spectra, the analyte amount can be assumed constant or nearly constant for NARCA and because models are formed for one sample, NARCA operates as a local modeling method. The performance of NARCA as a regression transfer learning method is evaluated using five near-infrared data sets.

Unsupervised model adaptation for multivariate calibration by domain adaptation-regularization based kernel partial least square

In chemometrics, calibration model adaptation is desired when training- and test-samples come from different distributions. Domain-invariant feature representation is currently a successful strategy to realize model adaptation and has received wide attention. The paper presents a nonlinear unsupervised model adaptation method termed as domain adaption regularization-based kernel partial least squares regression (DarKPLS). DarKPLS aims to minimize the source and target distributions in a low-dimensional latent space projected from the reproducing kernel Hilbert space (RKHS) generated with the labeled source data and unlabeled target data. Specially, the distributional means and variances between source and target latent variables are aligned in the RKHS. By extending existing domain invariant partial least square regression (di-PLS) with the projected maximum mean discrepancy (PMMD) to reduce the mean discrepancy in the RKHS further, DarKPLS could realize fine-grained domain alignment that further improves the adaptation performance. DarKPLS is applied to the γ-polyglutamic acid fermentation dataset, tobacco dataset and corn dataset, and it demonstrates improved prediction results in comparison with No adaptation partial least squares (PLS), null augmented regression (NAR), extended linear joint trained framework (ExtJT), scatter component analysis (SCA) and domain-invariant iterative partial least squares (DIPALS).

Feasibility study for transforming spectral and instrumental artifacts for multivariate calibration maintenance

Frequently, a spectral-based multivariate calibration model formed on a particular instrument (primary) needs to predict samples measured on other (secondary) instruments of the same spectral type. This situation is often referred to as calibration maintenance or transfer. A new calibration maintenance approach is developed in this paper using spectral differences between instruments. In conjunction with a sample weighting scheme, spectral differences are piecewise (wavelength window) or full spectrum fitted with modeling terms (correction terms) such as polynomials and derivatives. Results demonstrating the potential usefulness of the new method using a near infrared (NIR) benchmark dataset are presented in this paper. The process does not need a standardization sample set measured in the primary condition. Thus, the new approach is a "hybrid" between the popular methods of extended inverted multiplicative signal correction (EISC) and direct standardization (DS) or piecewise DS (PDS). It is found that prediction errors reduce for samples measured in the secondary condition compared to those based on no calibration transfer. Prediction errors are also comparable to those from a full calibration in the secondary condition. In addition to instrument correction, an extension of the new approach is discussed (but not tested) for predicting new samples changing over time due to new chemical, physical, and environmental measurement conditions including individually or combinations of temperature, sample particle size, and new spectrally responding species.

Food adulteration analysis without laboratory prepared or determined reference food adulterant values

Quantitative analysis of food adulterants is an important health and economic issue that needs to be fast and simple. Spectroscopy has significantly reduced analysis time. However, still needed are preparations of analyte calibration samples matrix matched to prediction samples which can be laborious and costly. Reported in this paper is the application of a newly developed pure component Tikhonov regularization (PCTR) process that does not require laboratory prepared or reference analysis methods, and hence, is a greener calibration method. The PCTR method requires an analyte pure component spectrum and non-analyte spectra. As a food analysis example, synchronous fluorescence spectra of extra virgin olive oil samples adulterated with sunflower oil is used. Results are shown to be better than those obtained using ridge regression with reference calibration samples. The flexibility of PCTR allows including reference samples and is generic for use with other instrumental methods and food products.