Calibration of Multiplexed Fiber-Optic Spectroscopy

Abstract

Large-scale commercial bioprocesses that manufacture biopharmaceutical products such as monoclonal antibodies generally involve multiple bioreactors operated in parallel. Spectra recorded during in situ monitoring of multiple bioreactors by multiplexed fiber-optic spectroscopies contain not only spectral information of the chemical constituents but also contributions resulting from differences in the optical properties of the probes. Spectra with variations induced by probe differences cannot be efficiently modeled by the commonly used multivariate linear calibration models or effectively removed by popular empirical preprocessing methods. In this study, for the first time, a calibration model is proposed for the analysis of complex spectral data sets arising from multiplexed probes. In the proposed calibration model, the spectral variations introduced by probe differences are explicitly modeled by introducing a multiplicative parameter for each optical probe, and then their detrimental effects are effectively mitigated through a "dual calibration" strategy. The performance of the proposed multiplex calibration model has been tested on two multiplexed spectral data sets (i.e., MIR data of ternary mixtures and NIR data of bioprocesses). Experimental results suggest that the proposed calibration model can effectively mitigate the detrimental effects of probe differences and hence provide much more accurate predictions than commonly used multivariate linear calibration models (such as PLS) with and without empirical data preprocessing methods such as orthogonal signal correction, standard normal variate, or multiplicative signal correction.

Quantitative analysis based on spectral shape deformation: A review of the theory and its applications

“…Accurate quantitative analysis of suspension samples was achieved by the combination of SSD with NIR spectroscopy . Other than the multiplicative light scattering effects, SSD can also solve the problems caused by the variation of sample's thickness and differences in the optical properties of NIR probes …”

Quantitative Analysis of Powder Mixtures by Raman Spectrometry: the influence of particle size and its correction

“…It accounts for the variations in Raman intensities caused by the changes in variables other than the mass fractions of the constituents in the powder mixtures, such as the intensity of the excitation source, the sample's particle size distribution, sample compactness, the overall mass and volume of the powder sample as well as the volume illuminated by the source and viewed by the spectrometer. , equation 4 can be rewritten as: Dual Calibration Strategy (DCS) 27,[29][30] For K training samples in which the mass fractions of the target constituent (say, the j-th constituent) are known, the multiplicative parameters, q k (k = 1, 2, …, K), can be estimated by the modified Optical Path-Length Estimation and Correction method (OPLEC m ) 30 ( the Matlab script for OPLEC m is provided in supporting information). After the estimation of q k (k = 1, 2, …, K), the following two calibration models can be built by multivariate linear calibration methods such as PLS.…”

Use of random forest in FTIR analysis of LDL cholesterol and tri‐glycerides for hyperlipidemia

“…The use of multivariate calibration methods is necessary to establish correlation to describe the change of concentration from these highly overlapped spectra. Multivariate techniques take the spectrum into account and exploit the multi‐channel nature of spectroscopic data to provide the signals from the spectral response of the analyte . Partial least squares (PLS) regression is the most routinely utilized in spectroscopic analytical applications, but a single PLS‐based classification method is hard to perform best for all different datasets in clinical diagnosis of blood plasma/serum because PLS models are always coupled with data colinearity.…”

“…Multivariate techniques take the spectrum into account and exploit the multichannel nature of spectroscopic data to provide the signals from the spectral response of the analyte. 32,33 Partial least squares (PLS) regression is the most routinely utilized in spectroscopic analytical applications, but a single PLS-based classification method is hard to perform best for all different datasets in clinical diagnosis of blood plasma/serum because PLS models are always coupled with data colinearity. Under this situation, some other chemometric methods have been also proposed.…”