Data-Driven Version of Multiway Soft Independent Modeling of Class Analogy (N-Way DD-SIMCA): Theory and Application.

One-class classification (OCC) is discussed in the framework of the measurement and processing of multiway data. Data-driven soft independent modeling of class analogy (DD-SIMCA) is applied in the following formats: (1) multiblock and (2) Tucker 3 N-way SIMCA, which are shown to be useful tools for solving classification tasks. A new decision rule for N-way DD-SIMCA is adopted based on the conventional two-way DD-SIMCA model. Multiblock SIMCA is shown to be useful for variable selection, and Tucker 3 SIMCA to select the optimal model complexity when applying multiway data decomposition and to assess the role of individual samples in the classification model. Both approaches, together with the two-way DD-SIMCA version applied to the unfolded data, are compared regarding the analysis of an experimental data set including genuine and adulterated blueberry extract samples. The latter were employed to produce matrix spectral-time data matrices per sample within a flow injection system, taking advantage of the spectral changes in the sample constituents as a function of the pH of the carrier phase. The need to employ the Tucker 3 model instead of a trilinear decomposition is supported by a discussion on the lack of the trilinearity property of the studied data.

Machine learning-couched treatment algorithms tailored to individualized profile of patients with primary anterior chamber angle closure predisposed to the glaucomatous optic neuropathy.

Background: Primary angle closure glaucoma (PACG) is still one of the leading causes of irreversible blindness, with a trend towards an increase in the number of patients to 32.04 million by 2040, an increase of 58.4% compared with 2013. Health risk assessment based on multi-level diagnostics and machine learning-couched treatment algorithms tailored to individualized profile of patients with primary anterior chamber angle closure are considered essential tools to reverse the trend and protect vulnerable subpopulations against health-to-disease progression. Aim: To develop a methodology for personalized choice of an effective method of primary angle closure (PAC) treatment based on comparing the prognosis of intraocular pressure (IOP) changes due to laser peripheral iridotomy (LPI) or lens extraction (LE). Methods: The multi-parametric data analysis was used to develop models predicting individual outcomes of the primary angle closure (PAC) treatment with LPI and LE. For doing this, we suggested a positive dynamics in the intraocular pressure (IOP) after treatment, as the objective measure of a successful treatment. Thirty-seven anatomical parameters have been considered by applying artificial intelligence to the prospective study on 30 (LE) + 30 (LPI) patients with PAC. Results and data interpretation in the framework of 3P medicine: Based on the anatomical and topographic features of the patients with PAC, mathematical models have been developed that provide a personalized choice of LE or LPI in the treatment. Multi-level diagnostics is the key tool in the overall advanced approach. To this end, for the future application of AI in the area, it is strongly recommended to consider the following:Clinically relevant phenotyping applicable to advanced population screeningSystemic effects causing suboptimal health conditions considered in order to cost-effectively protect affected individuals against health-to-disease transitionClinically relevant health risk assessment utilizing health/disease-specific molecular patterns detectable in body fluids with high predictive power such as a comprehensive tear fluid analysis. Supplementary Information: The online version contains supplementary material available at 10.1007/s13167-023-00337-1.

Procrustes cross-validation of multivariate regression models.

A generalization of Procrustes Cross-Validation - recently introduced novel approach for validation of chemometric models - is proposed. The generalized approach is faster than its predecessor by several orders of magnitude and can be used for validation of a wider range of models. Furthermore, it provides new tools for exploring the heterogeneity of the dataset, quality of cross-validation splits, presence of outliers, etc. The paper describes methodological aspects of the generalized approach, based on using Procrustean rules, the mathematical background, as well as presents practical results obtained using real chemical datasets.

Comparison of Lens Extraction Versus Laser Iridotomy on Anterior Segment, Choroid, and Intraocular Pressure in Primary Angle Closure Using Machine Learning.

PRCIS: Treatment strategy of primary angle closure (PAC) is not clear due to the large number of clinical and anatomic-topographic parameters in PAC, influencing the treatment algorithm. Using the machine learning method DD-SIMCA, we justify the expediency of early lens extraction (LE) in PAC. PURPOSE: To compare the anatomic and functional efficacy of LE and laser peripheral iridotomy (LPI) in patients with PAC using Machine Learning. MATERIALS AND METHODS: This prospective study included 120 patients aged 41-80 years: 60 eyes with PAC, 30 with PAC suspects, and 30 with healthy eyes (control). Thirty PAC eyes with intraocular pressure (IOP) up to 30 mm Hg were treated using LE with intraocular lens implantation and 30 eyes with LPI. All subjects underwent Swept Source optical coherence tomography. We analyzed 35 parameters of each eye including the lens vault, the choroidal thickness, the anterior chamber angle, and iris specifications such as iris curvature. Considering the correlations between them, the machine learning method DD-SIMCA 1-class classification was applied: the proximity of each sample to the target class (control) was characterized by the total distance to it. RESULTS: After LE, IOP was significantly lower than after LPI ( P =0). Every third eye with PAC after LE reached the target class: specificity according to DD-SIMCA equals 0.67. This was not observed for the eyes after LPI: specificity equals 1.0. After LE, all parameters of the anterior chamber angle did not differ from the control (all P >0.05). After LPI, there was an increase in anterior chamber depth ( P =0) and a decrease in lens vault ( P =0), but results comparable to the control were achieved only for iris curvature ( P =1.000). CONCLUSION: The efficacy of LE in PAC is higher than LPI due to the better postoperative anterior chamber topography and lower IOP. This study lends further clinical and anatomic support to the emerging notion of LE as an effective treatment for PAC.

Kinetic Model of Diclofenac Degradation Developed Using Multivariate Curve Resolution Method.

This study presents the kinetic modeling of the natural long-term aging of the pharmaceutical substance as well as the intact tablets of Diclofenac. Datasets are collections of near-infrared spectra acquired from the intact tablets packed in plastic blisters and the spectra of the pure substance. Fresh samples and samples at different stages of degradation are analyzed. No methods of accelerated aging were applied. Multi-step application of MCR-ALS in its soft version followed by the kinetic modeling of the results helps to propose a generic degradation mechanism; which includes: a global kinetic model; approximations of the NIR spectra of the intermediate and product; rough estimates of rate constants. We study tablets in blister packs; exactly as they are presented in pharmacies; and this is important from a practical point of view.

Aerosol Dry Printing for SERS and Photoluminescence-Active Gold Nanostructures Preparation for Detection of Traces in Dye Mixtures.

We proposed a novel method of nanostructure preparation for observation of surface-enhanced Raman spectroscopy (SERS) and metal-enhanced fluorescence (MEF) based on the deposition of gold nanoparticles (GNPs) above the thin dye film by dry aerosol printing. We detected various enhanced SERS and MEF signals of films of malachite green (MG) and rhodamine B (RhB) mixtures, depending on the surface packing density of Au NPs on the strip, and found the optimum one to achieve the 3.5 × 105 SERS enhancement. It was shown that statistical methods of chemometrics such as projection on latent structures provided the opportunity to distinguish SERS of MG from 100 ppm RhB in a mixture, whereas separation of MEF signals are feasible even for a mixture of MG and 1 ppm RhB due to two-photon excitation.

Procrustes Cross-Validation of short datasets in PCA context.

We suggest using a new tool, Procrustes cross-validation, as an alternative to a regular cross-validation for short datasets where each sample is important and, therefore, cannot be removed in line with the conventional leave-one-out cross-validation procedure. The advantages of the new approach are demonstrated using two real-world examples: the first one contains discrete variables (chemical profiles). The second one is based on continuous data (spectra). The method is implemented in R and Matlab as a small procedure that any analyst can easily use.

Procrustes Cross-Validation-A Bridge between Cross-Validation and Independent Validation Sets.

In this paper, we propose a new approach for validation of chemometric models. It is based on k-fold cross-validation algorithm, but in contrast to conventional cross-validation, our approach makes it possible to create a new dataset, which carries sampling uncertainty estimated by the cross-validation procedure. This dataset, called a pseudo-validation set, can be used similar to an independent test set, giving a possibility to compute residual distances, explained variance, scores, and other results, which cannot be obtained in the conventional cross-validation. The paper describes theoretical details of the proposed approach and its implementation as well as presents experimental results obtained using simulated and real chemical datasets.

Detection of Outliers in Projection-Based Modeling.

Previously, we have introduced an approach for calculation of the full object distance in the frame of Principal Component Analysis that can be applied to data exploration and classification. Now, a similar approach has been developed for regression problems in which a total distance can be calculated for every sample in projection modeling. Based on the total distance, a threshold for outlier detection has been developed by means of a data-driven estimation of the degrees of freedom and scaling parameters for the partial distances in the projection models. A joint threshold is used as a basis for a sequential outlier detection procedure. The iterative nature of the procedure helps to overcome masking effect in outliers, and a backward step eliminates swamping effects. Two real examples are used for illustration. The first dataset represents capsules filled with specially prepared mixtures of an active pharmaceutical ingredient and a number of excipients. This dataset is used to illustrate the behavior of possible outliers in the regression model and their corresponding locations in the X- and XY-distance plots. The second dataset consists of spectra of 135 whole wheat samples used for the prediction of protein, gluten, and moisture content. This dataset is used for a demonstration of the step-by-step application of the sequential procedure for outlier detection.

Chemometric Authentication of Brazilian Coffees Based on Chemical Profiling.

In this work, different chemometric tools were compared to classify n = 26 conventional (CONV) and n = 19 organic (ORG) coffees from the main Brazilian producing regions based on the chemical composition, physicochemical properties, and antioxidant activity. Principal component analysis separated ORG and CONV coffees but the distinction among the producing regions of Brazilian coffee was not possible. Partial least squares discriminant analysis classified all ORG and CONV coffees in the external validation. Similarly, linear discriminant analysis was able to discriminate 100% and 81% of ORG and CONV coffees in the external validation, respectively, in which total phenolic content (TPC), ferric reducing antioxidant activity, and caffeic acid were the main discriminant variables. Overall 100% of samples from Paraná, Minas Gerais, and blended samples were correctly classified, where TPC, flavonoids, inhibition of lipid peroxidation, caffeic acid, pH, and soluble solids were the main discriminant variables. Support vector machines classified 95% ORG and 88% CONV, 100% Coffea arabica, and 88% and 78% coffees produced in São Paulo and Minas Gerais. k-Nearest neighbors was effective in distinguishing 100% CONV, 89% ORG, 100% coffees from São Paulo, and 100% C. arabica coffees. Overall, HPLC data and simple physicochemical parameters allied to chemometrics were effective in authenticating the cultivation system and the botanical origin of Brazilian coffees. PRACTICAL APPLICATION: Coffee adulteration is a serious problem in the food chain as some fraudsters replace coffee powder by other cheaper products. In the case of organic coffee, this scenario is even worse as still there is not a universal method to differentiate conventionally grown coffee from its organic counterpart. In addition, Brazilian coffee is produced in different regions and the commercial value varies. Therefore, we analyzed some physicochemical, chemical, and antioxidant properties of Brazilian coffees from distinct origins and classified the samples using chemometrics. Our approach seems to be interesting for quality control purposes.

Chemometrics in analytical chemistry-part II: modeling, validation, and applications.

The contribution of chemometrics to important stages throughout the entire analytical process such as experimental design, sampling, and explorative data analysis, including data pretreatment and fusion, was described in the first part of the tutorial "Chemometrics in analytical chemistry." This is the second part of a tutorial article on chemometrics which is devoted to the supervised modeling of multivariate chemical data, i.e., to the building of calibration and discrimination models, their quantitative validation, and their successful applications in different scientific fields. This tutorial provides an overview of the popularity of chemometrics in analytical chemistry.

Trends in Chemometrics: Food Authentication, Microbiology, and Effects of Processing.

In the last decade, the use of multivariate statistical techniques developed for analytical chemistry has been adopted widely in food science and technology. Usually, chemometrics is applied when there is a large and complex dataset, in terms of sample numbers, types, and responses. The results are used for authentication of geographical origin, farming systems, or even to trace adulteration of high value-added commodities. In this article, we provide an extensive practical and pragmatic overview on the use of the main chemometrics tools in food science studies, focusing on the effects of process variables on chemical composition and on the authentication of foods based on chemical markers. Pattern recognition methods, such as principal component analysis and cluster analysis, have been used to associate the level of bioactive components with in vitro functional properties, although supervised multivariate statistical methods have been used for authentication purposes. Overall, chemometrics is a useful aid when extensive, multiple, and complex real-life problems need to be addressed in a multifactorial and holistic context. Undoubtedly, chemometrics should be used by governmental bodies and industries that need to monitor the quality of foods, raw materials, and processes when high-dimensional data are available. We have focused on practical examples and listed the pros and cons of the most used chemometric tools to help the user choose the most appropriate statistical approach for analysis of complex and multivariate data.

Chemometrics in analytical chemistry-part I: history, experimental design and data analysis tools.

Chemometrics has achieved major recognition and progress in the analytical chemistry field. In the first part of this tutorial, major achievements and contributions of chemometrics to some of the more important stages of the analytical process, like experimental design, sampling, and data analysis (including data pretreatment and fusion), are summarised. The tutorial is intended to give a general updated overview of the chemometrics field to further contribute to its dissemination and promotion in analytical chemistry.

Diffuse Reflectance Spectroscopy of Hidden Objects. Part II: Recovery of a Target Spectrum.

In this study, we consider the reconstruction of a diffuse reflectance near-infrared spectrum of an object (target spectrum) in case the object is covered by an interfering absorbing and scattering layer. Recovery is performed using a new empirical method, which was developed in our previous study. We focus on a system, which consists of several layers of polyethylene (PE) film and underlayer objects with different spectral features. The spectral contribution of the interfering layer is modeled by a three-component two-parameter multivariate curve resolution (MCR) model, which was built and calibrated using spectrally flat objects. We show that this model is applicable to real objects with non-uniform spectra. Ultimately, the target spectrum can be reconstructed from a single spectrum of the covered target. With calculation methods, we are able to recover quite accurately the spectrum of a target even when the object is covered by 0.7 mm of PE.

Diffuse Reflectance Spectroscopy of Hidden Objects, Part I: Interpretation of the Reflection-Absorption-Scattering Fractions in Near-Infrared (NIR) Spectra of Polyethylene Films.

Investigation of a sample covered by an interfering layer is required in many fields, e.g., for process control, biochemical analysis, and many other applications. This study is based on the analysis of spectra collected by near-infrared (NIR) diffuse reflectance spectroscopy. Each spectrum is a composition of a useful, target spectrum and a spectrum of an interfering layer. To recover the target spectrum, we suggest using a new phenomenological approach, which employs the multivariate curve resolution (MCR) method. In general terms, the problem is very complex. We start with a specific problem of analyzing a system, which consists of several layers of polyethylene (PE) film and underlayer samples with known spectral properties. To separate information originating from PE layers and the target, we modify the system versus both the number of the PE layers as well as the reflectance properties of the target sample. We consider that the interfering spectrum of the layer can be modeled using three components, which can be tentatively called transmission, absorption, and scattering contributions. The novelty of our approach is that we do not remove the reflectance and scattering effects from the spectra, but study them in detail aiming to use this information to recover the target spectrum.

Kinetic analysis of non-isothermal solid-state reactions: multi-stage modeling without assumptions in the reaction mechanism.

A novel non-linear regression method for modeling non-isothermal thermogravimetric data is proposed. Experiments for several heating rates are analyzed simultaneously. The method is applicable to complex multi-stage processes when the number of stages is unknown. Prior knowledge of the type of kinetics is not required. The main idea is a consequent estimation of parameters when the overall model is successively changed from one level of modeling to another. At the first level, the Avrami-Erofeev functions are used. At the second level, the Sestak-Berggren functions are employed with the goal to broaden the overall model. The method is tested using both simulated and real-world data. A comparison of the proposed method with a recently published 'model-free' deconvolution method is presented.

Spectrophotometric determination of Rare Earth Elements in aqueous nitric acid solutions for process control.

Noninvasive analytical control is of special interest for the complicated and hazardous production processes. On-line monitoring provides a unique opportunity to determine critical concentrations rapidly and without serious risks to operating personnel and the environment. Models for quantitative determination of concentrations of Rare Earth Elements in complex mixtures in nitric acid serve for these purposes. Here, the feasibility of simultaneous determination of cerium, praseodymium, and neodymium using the whole UV-vis spectroscopic range, together with chemometric data processing, is studied. The predictability of two chemometric techniques, partial least squares regression and correlation constrained multivariate curve resolution-alternating least squares are compared. Models' performances are analyzed in out-of-control cases.

The influence of fiber-probe accessories application on the results of near-infrared (NIR) measurements.

When several near-infrared instruments are used in a network and a common chemometric model is applied to spectral processing, comparison of the instruments is indispensable. Direct transferability often claimed by the producers should be treated with caution. It has been found experimentally that when measurements are performed with the help of a fiber optic probe, the main source of spectral discrepancy is related to probe sensitivity in contactless measurements. Here the influence of the probe-to-object distance on the acquired spectra is analyzed in detail. Special experimental setups are proposed to isolate various strongly influencing factors and to maintain stable measurement conditions. The application of an artificial standard instead of real-world objects helps to focus on the instrument/accessory characteristics.

In-line prediction of drug release profiles for pH-sensitive coated pellets.

A new method for the prediction of the drug release profiles during a running pellet coating process from in-line near infrared (NIR) measurements has been developed. The NIR spectra were acquired during a manufacturing process through an immersion probe. These spectra reflect the coating thickness that is inherently connected with the drug release. Pellets sampled at nine process time points from thirteen designed laboratory-scale coating batches were subjected to the dissolution testing. In the case of the pH-sensitive Acryl-EZE coating the drug release kinetics for the acidic medium has a sigmoid form with a pronounced induction period that tends to grow along with the coating thickness. In this work the autocatalytic model adopted from the chemical kinetics has been successfully applied to describe the drug release. A generalized interpretation of the kinetic constants in terms of the process and product parameters has been suggested. A combination of the kinetic model with the multivariate Partial Least Squares (PLS) regression enabled prediction of the release profiles from the process NIR data. The method can be used to monitor the final pellet quality in the course of a coating process.