Class modelling by Soft Independent Modelling of Class Analogy: why, when, how? A tutorial

Organic residues on equipment surfaces in poultry processing plants can generate cross contamination and increase the risk of unsafe food for consumers. This research was aimed to investigate the potential of LED-induced fluorescence imaging technique for rapid inspection of organic residues on poultry processing equipment surfaces. High-power blue LEDs with a spectral output at 400 nm were used as the excitation source for a line-scanning hyperspectral imaging system. Common chicken residue samples including fat, blood, and feces from ceca, colon, duodenum, and small intestine were prepared on stainless steel sheets. Fluorescence emission images were acquired from 120 samples (20 for each type of residue) in the wavelength range of 500 to 700 nm. LED-induced fluorescence characteristics of the tested samples were determined. PCA (principal component analysis) was performed to analyze fluorescence spectral data. Two SIMCA (soft independent modeling of class analogy) models were developed to differentiate organic residues and stainless steel samples. Classification accuracies using 2-class ('stainless steel' and 'organic residue') and 4-class ('stainless steel,' 'fat,' 'blood,' and 'feces') SIMCA models were 100% and 97.5%, respectively. An optimal single-band and a band-pair that are promising for rapid residue detection were identified by correlation analysis. The single-band approach using the selected wavelength of 666 nm could generate false negative errors for chicken blood inspection. Two-band ratio images using 503 and 666 nm (F503/F666) have great potential for detecting various chicken residues on stainless steel surfaces. This wavelength pair can be adopted for developing a LED-based hand-held fluorescence imaging device for inspecting poultry processing equipment surfaces.

Traditional analytical methods applied to the measurement of grape maturity and quality index in order to assess optimal harvest time have been proved to be slow and destructive. Therefore, non-destructive analytical techniques, including spectroscopy, can be a valid support for the choice of the best time to harvest. This study evaluated the feasibility of using a visible and near infrared spectral scanner (v. 1.4; DV Srl, Padova, Italy) with a detector in the region between 400-1000 nm to discriminate between grapes harvested at different times. Twelve clusters were harvested at 5 different times between October and December 2011. Spectra were acquired with a Spectral scanner on 3 intact berries from each bunch. These were randomly selected from top, medium and bottom zones, for a total of 180 spectra. Classification models were construed comparing 2 methods: soft independent modelling of class analogy (SIMCA) and partial least squares discriminant analysis (PLS-DA). The SIMCA model was developed building individual principal component analysis (PCA) models for the spectra of each harvest time. Different pre-treatment methods were tested in order to enhance the power of the model, thus enhancing the score differences among samples from different harvest times. The transformation that allowed the best statistical separation among scores of grapes from different harvest times was the second derivate of Norris. Therefore, the PCA model obtained from the spectra subjected to this pre-treatment was used for SIMCA classification. The PLS-DA model were developed applying the PLS2 algorithm. In order to construct discriminant models to classify bunch spectra according to the 5 harvest times, spectral variations were correlated with the 5 categories established. No pretreatments were previously applied in this last case since they did not improve the final result. The SIMCA method was unable to correctly classify grapes from harvest time 2 (59% of correct classification) and was less efficient compared to the PLS-DA model. Using the PLS-DA model, all the grapes were correctly classified (100%) with the exception of those from harvest time 5 (94%). The overall results demonstrate that this method has excellent potential for discriminating grape quality.

Traditional analytical methods applied to the measurement of grape maturity and quality index in order to assess optimal harvest time have been proved to be slow and destructive. Therefore, non-destructive analytical techniques, including spectroscopy, can be a valid support for the choice of the best time to harvest. This study evaluated the feasibility of using a visible and near infrared spectral scanner (v. 1.4; DV Srl, Padova, Italy) with a detector in the region between 400-1000 nm to discriminate between grapes harvested at different times. Twelve clusters were harvested at 5 different times between October and December 2011. Spectra were acquired with a Spectral scanner on 3 intact berries from each bunch. These were randomly selected from top, medium and bottom zones, for a total of 180 spectra. Classification models were construed comparing 2 methods: soft independent modelling of class analogy (SIMCA) and partial least squares discriminant analysis (PLS-DA). The SIMCA model was developed building individual principal component analysis (PCA) models for the spectra of each harvest time. Different pre-treatment methods were tested in order to enhance the power of the model, thus enhancing the score differences among samples from different harvest times. The transformation that allowed the best statistical separation among scores of grapes from different harvest times was the second derivate of Norris. Therefore, the PCA model obtained from the spectra subjected to this pre-treatment was used for SIMCA classification. The PLS-DA model were developed applying the PLS2 algorithm. In order to construct discriminant models to classify bunch spectra according to the 5 harvest times, spectral variations were correlated with the 5 categories established. No pretreatments were previously applied in this last case since they did not improve the final result. The SIMCA method was unable to correctly classify grapes from harvest time 2 (59% of correct classification) and was less efficient compared to the PLS-DA model. Using the PLS-DA model, all the grapes were correctly classified (100%) with the exception of those from harvest time 5 (94%). The overall results demonstrate that this method has excellent potential for discriminating grape quality.

We report soft independent modeling of class analogy (SIMCA) analysis of laser-induced plasma emission spectra of edible salts from 12 different geographical origins for their classification model. The spectra were recorded by using a simple laser-induced breakdown spectroscopy (LIBS) device. Each class was modeled by principal component analysis (PCA) of the LIBS spectra. For the classification of a separate test data set, the SIMCA model showed 97% accuracy in classification. An additional insight could be obtained by comparing the SIMCA classification result with that of partial least squares discriminant analysis (PLS-DA). Different from SIMCA, the PLS-DA classification accuracy seems to be sensitive to addition of new sample classes to the whole data set. This indicates that the individual modeling approach (SIMCA) can be an alternative to global modeling (PLS-DA), particularly for the classification problems with a relatively large number of sample classes.

The potential of near-infrared (NIR) spectroscopy to determine the geographical origin of honey samples was evaluated. In total, 167 unfiltered honey samples (88 Irish, 54 Mexican, and 25 Spanish) and 125 filtered honey samples (25 Irish, 25 Argentinean, 50 Czech, and 25 Hungarian) were collected. Spectra were recorded in transflectance mode. Following preliminary examination by principal component analysis (PCA), modeling methods applied to the spectral data set were partial least-squares (PLS) regression and soft independent modeling of class analogy (SIMCA); various pretreatments were investigated. For unfiltered honey, best SIMCA models gave correct classification rates of 95.5, 94.4, and 96% for the Irish, Mexican, and Spanish samples, respectively; PLS2 discriminant analysis produced a 100% correct classification for each of these honey classes. In the case of filtered honey, best SIMCA models produced correct classification rates of 91.7, 100, 100, and 96% for the Argentinean, Czech, Hungarian, and Irish samples, respectively, using the standard normal variate (SNV) data pretreatment. PLS2 discriminant analysis produced 96, 100, 100, and 100% correct classifications for the Argentinean, Czech, Hungarian, and Irish honey samples, respectively, using a second-derivative data pretreatment. Overall, while both SIMCA and PLS gave encouraging results, better correct classification rates were found using PLS regression.

Application of chemometrics for detection and modeling of adulteration of fresh cow milk with reconstituted skim milk powder using voltammetric fingerpriting on a graphite/ SiO2 hybrid electrode

Application of hand-held near-infrared and Raman spectrometers in surface treatment authentication of cork stoppers

Nondestructive estimation of maturity and textural properties on tomato ‘Momotaro’ by near infrared spectroscopy

ABSTRACTUltraviolet-visible (UV-Vis) spectral information (190–400 nm) was used to classify coffee samples into either pure peaberry or pure normal coffee classes using two chemometric methods: soft independent modelling of class analogy (SIMCA) and partial least squares discriminant analysis (PLS-DA). The spectral data of peaberry and normal coffee were acquired using a UV-Vis spectrometer (Genesys™ 10S UV-Vis, Thermo Scientific, USA). For both supervised discrimination methods, SIMCA and PLS-DA, all samples were correctly classified into their corresponding classes. The SIMCA model classified all samples accurately (100%) into either the peaberry or normal coffee class, even at a 5% confidence level; however, the PLS-DA model also correctly classified all samples (100%). Investigation of the major wavelengths contributing to the classifications using x-loading weights and loading of latent variables (LVs) indicated wavelengths at 230, 250, 270, 310, and 350 nm were important for determination of the coffee types. These wavelengths were closely related to the absorbance wavelengths of several important chemical components in roasted coffee: caffeine, caffeic acid, and chlorogenic acids (CGA). These results provided the basis for developing a simple and reliable method for peaberry coffee authentication (including more complex samples where peaberry coffee is blended with normal coffees) based on UV-Vis spectra.

The selection of suitable rice varieties is the key to achieve high and stable yields, and the correct identification of rice varieties is the prerequisite for seed selection. In this paper, with Kenjing No.5, No.6, and No.9 as the subjects, the effectiveness of near-infrared spectroscopy (NIRS) combined with soft independent modeling of class analogy (SIMCA) in the rapid identification of rice varieties was explored. The modeling sets of Kenjing No.5, No.6, and No.9 samples were respectively used to establish a SIMCA classification model based on principal component analysis (PCA). The accuracies of the model in classifying the rice samples in the modeling set were 100, 100, and 97.5%, respectively. Then, the established SIMCA model was used to identify the rice samples in the test set. According to the experimental findings, the SIMCA analytical method achieved 100% prediction accuracy for the Kenjing No.5, Kenjing No.6, and Hongyu 001-1 samples. For the Kenjing No.9 sample, the accuracy rate was 90% with a 10% sample of Kenjing No.9 misidentified as Kenjing No.6. Therefore, the analytical method of NIRS combined with SIMCA could effectively identify the rice varieties, providing a new approach for the correct selection of planting varieties.

Evaluation of single and multi-feedstock biodiesel – diesel blends using GCMS and chemometric methods

Detection and treatment of arrhythmias has become one of the main goals in cardiac care diagnosis provided by general practitioners. Electrocardiogram (ECG) analysis is one of the most commonly used tools to test and diagnose heart problems. Classification of ECG heartbeats enables the identification of specific arrhythmia or other heart conditions. This paper presents and contrasts the results from two effective ECG arrhythmia classification schemes. The first scheme consists of a principal component analysis (PCA) step for feature reduction at the input vector to the classifier, combined with soft independent modelling of class analogy (SIMCA). The second method uses a multi-class support vector machine (MSVM) classifier to differentiate between four different types of arrhythmia from ECG beats. The four types of beats include Normal (N), Premature Ventricular Contraction (PVC), and Atrial premature contraction (APC) and Right Bundle Branch Block Beat (RBBB). The time domain features were obtained from the St Petersburg INCART 12-lead Arrhythmia Database (incartdb). Between 10 and 30 Principal Components (PCs) were selected for reconstructing individual ECG beats and create the input vector to the classifier. The average classification accuracy of the proposed scheme is 76.83% and 98.33% using MSVM and SIMCA classifier respectively. The SIMCA classification algorithm provided better performance than the MSVM classifiers.

Source- and age-related biomarker and isotopic data were measured for more than 1000 crude oil samples from wells and seeps collected above approximately 55°N latitude. A unique, multitiered chemometric (multivariate statistical) decision tree was created that allowed automated classification of 31 genetically distinct circum-Arctic oil families based on a training set of 622 oil samples. The method, which we call decision-tree chemometrics, uses principal components analysis and multiple tiers of K-nearest neighbor and SIMCA (soft independent modeling of class analogy) models to classify and assign confidence limits for newly acquired oil samples and source rock extracts. Geochemical data for each oil sample were also used to infer the age, lithology, organic matter input, depositional environment, and identity of its source rock. These results demonstrate the value of large petroleum databases where all samples were analyzed using the same procedures and instrumentation.

The present study proposes a variable selection algorithm for Soft Independent Modeling of Class Analogy (SIMCA) in the context of rigorous one-class classification (OCC), using Modeling Power (MP) as a central internal metric. The algorithm integrates three complementary criteria: correlation between MP and class compactness, MP non-growth rate across successive principal components, and a minimum MP threshold. These criteria were applied to three experimental datasets: UV-Vis spectra of edible oils, NIR spectra of Argentinean green teas, and HPLC-CAD chromatographic profiles of olive oils. In all cases, the goal was to enhance model parsimony and interpretability without compromising classification performance. The selected variables were chemically meaningful, aligning with known spectral or chromatographic regions associated with key compositional markers. Comparative analyses between the proposed Modeling Power Selector with SIMCA (MPS-SIMCA) and traditional full-spectrum SIMCA showed equivalent or improved classification performance. Additionally, MPS-SIMCA achieved superior model compactness and interpretability, supporting the feasibility of variable selection based solely on internal class structure. This approach offers a robust and interpretable alternative for class modeling in food authentication tasks where only target-class samples are reliably available.

UV-Vis spectroscopy and one-class modeling for the authentication of the geographical origin of green coffee beans from Cerrado Mineiro, Brazil