Search

The monitoring and control of bioprocesses rely on the measurement of the main metabolite concentrations. To this end, infrared spectroscopy (IR) is a good candidate with which to perform rapid and non-destructive measurements. However, IR-based measurements rely on a calibration step linking the measured spectra to the concentrations of the compounds of interest. This calibration may suffer with problems of robustness when the measuring conditions change, such as when some chemicals not present in the calibration spectra are added when using the IR sensor. In this study, a method based on orthogonal projection, dynamic orthogonal projection (DOP), was tested for its ability to cope with the robustness problem caused by the addition of ammonia in a pilot-scale anaerobic digester, whose volatile fatty acid concentrations were monitored by mid-IR spectrometry. The results demonstrate that DOP has significant potential as a form of process analytical technology. [ABSTRACT FROM AUTHOR]

Differences in the relative spectral response functions of sensors lead to data inconsistencies that should be harmonized before multisensor exploitation. In this paper, we use spectral libraries to simulate satellite data and build models to correct them. We then explore and compare different models for coarse and medium spatial resolution optical sensors, including moderate resolution imaging spectroradiometer, advanced very high resolution radiometer (AVHRR), visible infrared imaging radiometer suite, multispectral instrument aboard Sentinel-2, and Operational Land Imager aboard Landsat 8. We found that optimal correction of different bands depends on the model used. For the green and near infrared bands, a multilinear land cover dependent regression improves the accuracy by up to 80.9%. For the red band, a novel exponential dependence of the spectral band adjustment factor with the normalized difference vegetation index (NDVI) provides an accuracy improvement of up to 72.8%. The best way to correct the NDVI value is to use the corrected NIR and red bands using these models. We apply the proposed methods to 445 BELMANIP2 sites using AVHRR data from the long-term data record from 1982–2017. High NDVI pixels result in 30-year trends varying up to 0.06 when comparing uncorrected to spectrally adjusted NDVI. Further application of these methods to NASA's Harmonized Landsat and Sentinel 2 product shows that for the red band and NDVI, our proposed method provides improved accuracy (54.6% and 62.5%) over the linear spectral adjustment currently used. [ABSTRACT FROM AUTHOR]

Calibration transfer has been traditionally performed in the context of transferring models between instruments using standard samples. Recently, new methodologies and applications have shown that transfer techniques can be adopted to achieve calibration transfer between other types of domains, such as product form, variant or seasonality. In addition, to achieving a higher efficiency for calibration transfer, it is desirable to perform the transfer without the need for standard samples or new reference analyses. Therefore, we propose a method for unsupervised calibration transfer based on the orthogonalization for structural differences between domains. The method has been successfully applied to one simulated dataset and two real datasets. In the studied cases, the proposed methodology allowed to achieve a successful transfer of calibration models and enabled the interpretation of the interferences responsible for the degradation of the original calibration models when transferred to the new domain. [Display omitted] [ABSTRACT FROM AUTHOR]

Abstract: This article describes several linear pretreatments based on orthogonal projections. The main differences of these pretreatments lie in the way the information to be removed are identified, using calibration dataset, pure spectra, experimental designs or mathematical models. Removing all the undesired spectral information yields spectra proportional to the net analyte signal, so it is important to collect the most complete information possible, using the complementarities of different approaches. The correction should then be processed with a single Euclidian orthogonal projection that gathers all the information, rather than with successive operations. By embedding Euclidian orthogonal projections into the calibration, it is not necessary to reapply them to new datasets. [Copyright &y& Elsevier]

Abstract: Several linear calibration methods have been proposed for predicting the concentration of a particular compound from a spectrum. Some methods are based on experimental data, such as Partial Least Square Regression. Other methods are based on expert data, e.g. Direct Calibration. This article proposes a new method, called Improved Direct Calibration, which uses expert and experimental information. It performs a projection onto the pure interest spectrum, after correcting it from influence factors. No calibration dataset is necessary to build this model. This method has been successfully applied to the quantification of ethanol in musts during fermentation, using near infrared spectrometry. [Copyright &y& Elsevier]

Ensemble pre-processing is emerging as a potential tool to avoid the tiring pre-processing selection and optimization task in near-infrared (NIR) spectral modelling. Furthermore, differently pre-processed data may carry complementary information, hence, ensemble pre-processing may represent the best suited modelling option to extract all the useful information from differently pre-processed data. Recently, multi-block techniques such as sequential (SPORT) and parallel (PORTO) orthogonalized partial least squares regression were proposed to extract complementary information present in differently pre-processed data. Although such multi-block techniques allowed efficient modelling of differently pre-processed data blocks, depending on the approach, challenges related to choosing block order, parameter tuning, block scaling and optimization time requirements still must be dealt with. To cope with such issues, the present study proposes the use of a recently developed faster, block order independent and scale independent, multi-block data modelling technique called response-oriented sequential alternation (ROSA) to process the multi-block data generated by differently pre-processing the same NIR data. This new method is called PROSAC, i.e., pre-processing ensembles with ROSA calibration. The potential of the approach is demonstrated on five real NIR spectral datasets. Furthermore, as baselines for comparison, partial least squares regression was done on individually pre-processed data sets, and using two multi-block pre-processing fusion approaches, i.e., SPORT and PORTO. The ensemble pre-processing with ROSA achieved either better performance compared to the baseline methods or achieved comparable performance without the need to worry about the pre-processing order, the scaling of data after pre-processing and optimization time requirements. PROSAC can be considered as a general tool for the ensemble pre-processing for NIR data modelling. • The method was compared with recent novel multi-block pre-processing ensemble tools. • The method is fast, order and block scale independent thus complements pre-processing ensembles. • The method was evaluated on five NIR spectroscopy datasets. • The method eliminates the need for pre-processing selection. [ABSTRACT FROM AUTHOR]

Effective methods often rely on simple mathematical operators. Among these operators, orthogonal projections have been widely used because of their simplicity in compensating for detrimental factors. This efficiency depends largely on the way these tools are prepared. This article links the mathematical basics of orthogonal projections to the notion of vectoral subspaces, highlighting which information should be removed in the process and the important practical properties concerned with optimizing this technique. This review covers several methods involving orthogonal projections and focuses specifically on their practical use. This concerns the identification of detrimental information and its removal together with adjusting the dimension of the projection. The methodology discussed in this review will enable the reader to optimize orthogonal projections for any given situation. The concept and importance of orthogonal projections are presented and situated within pretreatments and calibrations. The key points of orthogonal projections are noted: identifying the right information, then building a basis of the subspace to remove detrimental information. [ABSTRACT FROM AUTHOR]

Transmission Raman Spectroscopy (TRS) was implemented as an 'Extended' Content Uniformity (ECU) method for un-coated tablets for a commercial pharmaceutical product. By sampling un-coated tablets throughout the duration of the tablet compression stage, it can be demonstrated that the material from the preceding blend step was of uniform composition, and therefore the blend and compression unit-operations were in a state of control. TRS was selected as a rapid, non-destructive measurement that can be automated through the use of a sample tray that can hold many tablets. In this work, the performance of a multivariate calibration model (PLS) deployed to two Transmission Raman Spectrometers co-located within the same QC laboratory was studied using data obtained over a 3-year period. The aim of the investigation was to assess the impact of various annual instrument maintenance events, and to evaluate several chemometric methods for reducing or eliminating the spectral effects that led to deterioration of a models performance. Linear orthogonal projection approaches such as Transfer by Orthogonal Projection (TOP), Dynamic Orthogonal Projection (DOP) and Unsupervised Dynamic Orthogonal Projection (uDOP) were applied, along with a more recent, non-linear method called Transfer Component Analysis-Orthogonal Projection (TCA-OP). This works shows that each method has merits, depending on the nature of the spectral/model correction required. In most cases, the model performance could be fully restored, or significantly improved. This work also highlights how these various methods can be useful tools to better understand the root-cause for a deterioration in model performance. • Transmission Raman Spectroscopy calibration model was successfully validated and deployed to two co-located instruments for an established pharmaceutical product. • The impact of 3 successive, annual instrument service interventions were studied to evaluate the impact on model performance when no model update was performed. • A non-linear and several linear orthogonal projection methods were applied and compared. • In most cases, the model performance could be fully restored, or significantly improved. [ABSTRACT FROM AUTHOR]

Calibration transfer is an essential activity in analytical chemistry in order to avoid a complete recalibration. Currently, the most popular calibration transfer methods, such as piecewise direct standardization and dynamic orthogonal projection, require a certain amount of standard or reference samples to guarantee their effectiveness. To achieve higher efficiency, it is desirable to perform the transfer with as few reference samples as possible. To this end, we propose a new calibration transfer method by using a calibration database from a master instrument (source domain) and only one spectrum with known properties from a slave instrument (target domain). We first generate a counterpart of this spectrum in the source domain by a multivariate Gaussian kernel. Then, we train a filter to make the response function of the slave instrument equivalent to that of the master instrument. To avoid the need for labels from the target domain, we also propose an unsupervised way to implement our method. Compared with several state-of-the-art methods, the results on one simulated dataset and two real-world datasets demonstrate the effectiveness of our method. Traditionally, the demand for certain amounts of reference samples during calibration transfer is cumbersome. Our approach, which requires only one reference sample, makes the transfer process simple and fast. In addition, we provide an alternative for performing unsupervised calibration transfer. As such, the proposed method is a promising tool for calibration transfer. [Display omitted] • A new calibration transfer method is introduced, which requires only one spectrum from target domain. • The calibration transfer method can be conducted in an unsupervised way. • A multivariate Gaussian kernel is introduced to generate a virtual sample in source domain. • Results from simulated and real-world datasets demonstrate the effectiveness of our method. [ABSTRACT FROM AUTHOR]

A general model is proposed for the explanation of several linear methods used for quantification. A first projection approach concerns pretreatments and calibrations. It consists of an oblique projection of the spectral data onto a subspace containing useful information for calibrations or detrimental information for pretreatments. Corrected spectra and scores are obtained for pretreatments and calibrations, respectively. A second projection approach concerns only calibrations. The regression vector is deduced after an orthogonal projection of the reference values onto the scores previously obtained. Several pretreatments, and direct and indirect (inverse) calibrations also called regressions are reviewed according to this model. The methods described are focused on spectroscopic applications. Some are very specific to spectroscopy; however, most of them also can be applied in other situations. [ABSTRACT FROM AUTHOR]

Abstract: The standard PLSR is presented from a geometric point of view consisting of two projections. In the first, the scores are obtained after an oblique projection of the spectra onto the loadings. In the second, the vector of response values is projected orthogonally onto the scores. A metric is introduced for the oblique projection and a new algorithm for the calculation of the loadings into the variables space is proposed. This work also develops a new parameter, a vector, whose different values lead to different regression models with their own abilities of prediction; one of them is the exact form of the standard PLSR. Two applications are described to illustrate the performance of the proposed method called VODKA regression, which is also a way to build least square regressions by introducing additional knowledge into the models. [Copyright &y& Elsevier]

Abstract: Following up on the success of previous chemometric challenges arranged during the annual congress organised by the French Chemometrics Society, the organisation committee decided to repeat the idea for the Chimiometrie 2007 event (http://www.chimiometrie.org/) held in Lyon, France (29–30 November) by featuring another dataset on its website. As for the first contest in 2004, this dataset was selected to test the ability of participants to apply regression methods to NIR data. The aim of Challenge 2007 was to perform a calibration model as robust and precise as possible using a data set with only a few reference values available and submitted to different perturbation factors. The committee received nine answers; this paper summarizes the best three approaches, as well as the approach proposed by the organisers. [Copyright &y& Elsevier]

Abstract: Near-infrared (NIR) and mid-IR spectroscopy applied to soil compositional analysis started to develop markedly in the 1990s, taking advantage of earlier advances in instrumentation and chemometrics for agricultural products. Today, NIR spectroscopy is envisioned as replacing laboratory analysis in certain applications (e.g., soil-carbon-credit assessment at the farm level). However, accuracy is still unsatisfactory compared with standard laboratory procedures, leading some authors to think that such a challenge will never be met. This article investigates the critical points to be aware of when accuracy of NIR-based measurements is assessed. First is the decomposition of the standard error of prediction into components of bias and variance, only the latter being reducible by averaging. This decomposition is not used routinely in the soil-science literature. Contrarily, a log-normal distribution of reference values is very often encountered with soil samples [e.g., elemental concentrations (e.g., carbon)] with numerous small or zero values. These very skewed distributions make us take precautions when using inverse regression methods (e.g., principal component regression or partial least squares), which force the predictions towards the centre of the calibration set, leading to negative effects on the standard error prediction – and therefore on prediction accuracy – especially when log-normal distributions are encountered. Such distributions, which are very common for soil components, also make the ratio of performance to deviation a useless, even hazardous, tool, leading to erroneous conclusions. We propose a new index based on the quartiles of the empirical distribution – ratio of performance to inter-quartile distance – to overcome this problem. [Copyright &y& Elsevier]

Calibration transfer (CT) refers to the set of chemometric techniques used to transfer (near-infrared) calibration models between spectrometers. The requirement of traditional CT methods to measure calibration standard samples has been a challenge as such measurements are difficult in real-world applications, e.g. when the instruments are located far apart or chemically stable standard samples are not available. In recent years, major developments have taken place in the domain of CT, hence, this work provides a concise but critical review of all the main recent chemometric techniques available to perform CT. Particularly this work explains some newer concepts for standard-free CT, where the standard samples are not required to attain the CT. We conclude that CT approaches that do not rely on standard sample measurements hold promise to help making calibration models sharable between similar analytical devices and to increase the applicability of CT to real-world problems in the analytical sciences. • All major calibration transfer (CT) methods are summarised. • New CT approaches are critically reviewed. • Standard-free CT techniques are summarised. • Out of box application of CT methods are summarised. [ABSTRACT FROM AUTHOR]