Supervised Pattern Recognition Techniques Research Articles

Discrimination/Classification of Edible Vegetable Oils from Raman Spatially Solved Fingerprints Obtained on Portable Instrumentation.

Currently, the combination of fingerprinting methodology and environmentally friendly and economical analytical instrumentation is becoming increasingly relevant in the food sector. In this study, a highly versatile portable analyser based on Spatially Offset Raman Spectroscopy (SORS) obtained fingerprints of edible vegetable oils (sunflower and olive oils), and the capability of such fingerprints (obtained quickly, reliably and without any sample treatment) to discriminate/classify the analysed samples was evaluated. After data treatment, not only unsupervised pattern recognition techniques (as HCA and PCA), but also supervised pattern recognition techniques (such as SVM, kNN and SIMCA), showed that the main effect on discrimination/classification was associated with those regions of the Raman fingerprint related to free fatty acid content, especially oleic and linoleic acid. These facts allowed the discernment of the original raw material used in the oil's production. In all the models established, reliable qualimetric parameters were obtained.

Application of linear and nonlinear pattern recognition techniques for discrimination of fluoroquinolones using modified AgNPs-based colorimetric sensor array

Surface plasmon resonance-based colorimetric sensor arrays made up of metallic nanoparticles (NPs) are powerful tools that are mostly used for pattern recognition and quantification purposes, especially in the case of pharmaceutical and biologically active substances. Ease of fabrication, simplicity, cost-effectiveness, and huge amounts of collectible data besides the various fabrication techniques, expand their applications in analytical chemistry, especially when combined with chemometrics data analysis tools. In this study, a colorimetric sensor array is fabricated using unmodified citrate-capped together with the modified AgNPs for the discrimination of three fluoroquinolone drugs ofloxacin, ciprofloxacin, and moxifloxacin in real serum samples. The aggregation-induced color changes of the AgNPs were monitored during the time and the collected kinetic-spectrophotometric huge data in the ranges 300–800 nm was compressed by Discrete Wavelet Transform (DWT) to remove the redundant variables and analyzed with different unsupervised and supervised pattern recognition techniques like Principal Components Analysis (PCA), Linear Discriminant Analysis (LDA) and XY-Fused neural networks (XYF). The best classification performance was achieved using the genetic algorithm-optimized XY-Fused networks with 92% accuracy and an error of cross-validation equal to zero. In addition, all of the spiked real serum samples were classified correctly using the generated model according to the assigned class labels.

Comparison of Geographical Traceability of Wild and Cultivated Macrohyporia cocos with Different Data Fusion Approaches.

Poria originated from the dried sclerotium of Macrohyporia cocos is an edible traditional Chinese medicine with high economic value. Due to the significant difference in quality between wild and cultivated M. cocos, this study aimed to trace the origin of the fungus from the perspectives of wild and cultivation. In addition, there were quite limited studies about data fusion, a potential strategy, employed and discussed in the geographical traceability of M. cocos. Therefore, we traced the origin of M. cocos from the perspectives of wild and cultivation using multiple data fusion approaches. Supervised pattern recognition techniques, like partial least squares discriminant analysis (PLS-DA) and random forest, were employed in this study using. Five types of data fusion involving low-, mid-, and high-level data fusion strategies were performed. Two feature extraction approaches including the selecting variables by a random forest-based method—Boruta algorithm and producing principal components by the dimension reduction technique of principal component analysis—were considered in data fusion. The results indicate the following: (1) The difference between wild and cultivated samples did exist in terms of the content analysis of vital chemical components and fingerprint analysis. (2) Wild samples need data fusion to realize the origin traceability, and the accuracy of the validation set was 95.24%. (3) Boruta outperformed principal component analysis (PCA) in feature extraction. (4) The mid-level Boruta PLS-DA model took full advantage of information synergy and showed the best performance. This study proved that both geographical traceability and optimal identification methods of cultivated and wild samples were different, and data fusion was a potential technique in the geographical identification.

Chemometric Approaches in Questioned Documents

Questioned documents comprehend analysis of identity theft, forged signatures or texts, documents alterations and falsification of security documents or banknotes. Questions involving inks or paper require chemical analysis, and multivariate analysis or chemometrics has been an emerging tool for data evaluation and interpretation after instrumental data collection in this area. The goal of this study is to identify previous articles that applied multivariate analysis within questioned documents for forensic purposes. The search for articles was performed in four databases (Google Scholar, Science Direct, Pubmed and Scopus). Sixty studies, published in the last ten years, were selected. Thirty-four articles described pen inks analysis; fourteen articles comprehended printed documents studies; eight articles evaluated paper analysis, and four articles included banknotes analysis. Spectroscopy, mass spectrometry, chromatography, thermo gravimetric analysis and multivariate image analysis were the analytical methods applied to collect chemical data. Chemometrics methods included mainly unsupervised pattern recognition techniques, regression methods, and supervised pattern recognition techniques, amongst other methods. This review summarized and discussed multivariate analysis techniques applied in different questioned documents sub-areas, highlighting the importance of this knowledge for forensic analysts. In addition, it shows new research topics such as different printing and pen inks, papers and security documents analysis herein not included.

Infrared spectroscopic and chemometric approach for identifying morphology in embryo culture medium samples

The present study aimed to examine embryo culture medium (ECM) samples in in vitro fertilization (IVF) using Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy coupled with the pattern recognition methods in order to differentiate between morphology A, morphology AB and morphology B results.In this work, 45 the ECM samples analyzed in the 1100–3000 cm−1 spectral region. The FTIR data from ECM samples were subjected to multivariate analyses by the unsupervised and supervised pattern recognition techniques. Cluster analysis (CA) after diagnosis and elimination outlier detection was performed to process FTIR data. Also, partial least squares discriminant analysis (PLS-DA) as a parametric and linear supervised classification algorithm and K-nearest neighbors (KNN) as a non-parametric and linear supervised classification algorithm were performed in spectral data analysis for the classification. The supervised classification algorithms were also tested via the variable selection by means of genetic algorithm (GA).The classification efficiency parameters, including accuracy (ACC), error rate (ER) and non-error rate (NER) were calculated and finally, sum of the ranking differences (SRD) procedure was used to compare chemometrics methods.The GA-PLS-DA approach with orthogonal signal correction (OSC) preprocessing (Case 4) for classification of the ECM samples exhibited the best performance and obtained ACC, ER, NER of 91%, 4% and 96% in the test set, respectively.Hence, the proposed method is rapid, simple, without any chemical preparation and accurate for the discrimination of the ECM samples based on their morphology results. In addition, the SRD procedure was used to order and group chemometric methods applied for the discrimination of the ECM samples based on their morphology results.

Identification of waxy cassava genotypes using fourier‐transform near‐infrared spectroscopy

AbstractHigh‐throughput phenotyping tools that allow the early and accurate evaluation of important agronomic traits have gained space in current breeding programs. The aim of this study was to evaluate the potential of Fourier‐transform near‐infrared spectroscopy (FT‐NIRS) to identify cassava (Manihot esculenta Crantz) clones with waxy starch (i.e., amylose‐free) by screening leaves rather than roots, and to validate prediction models for classifying these phenotypes. We analyzed the spectra of 162 waxy and 180 nonwaxy genotypes from five different growing environments. The mean FT‐NIRS spectra and principal component analysis (PCA) were used to investigate the potential for grouping the data. For classification, five supervised pattern recognition techniques were tested: Bayesian generalized linear model (BGLM), high‐dimensional discriminant analysis (HDDA), partial least squares‐discriminant analysis (PLS‐DA), parallel random forest (PRANDF), and support vector machines with linear kernel (SVM). The mean spectra and the PCA did not allow discrimination of the genotypes based on starch classification. The SVM and BGLM showed the highest classification accuracy in cross‐validation (.86–.87), with higher concordance rates (.88–.83), sensitivity (.87–.85) and specificity (.88). The BGLM and SVM models also obtained better indices in the external validation, with high accuracy (.85) and correct classification of 93% of the waxy genotypes. Thus, performing early selection of root characteristics based on the indirect selection of variables extracted from leaf spectra is a good potential strategy for more efficient breeding of the waxy phenotype.

Pattern recognition-based Raman spectroscopy for non-destructive detection of pomegranates during maturity.

In this study, the feasibility of utilizing Fourier transform Raman spectroscopy, combined with supervised and unsupervised pattern recognition methods was considered, to distinguish the maturity stage of pomegranate "Ashraf variety" during four distinct maturity stages between 88 and 143days after full bloom. Principal component analysis (PCA) as an unsupervised pattern recognition method was performed to verify the possibility of clustering of the pomegranate samples into four groups. Two supervised pattern recognition techniques namely, partial least squares Discriminant Analysis (PLS-DA) and Soft Independent Modeling of Class Analogy (SIMCA) were also used as powerful supervised pattern recognition methods to classify the samples. The results showed that in all groups of samples, the Raman spectra of the samples were correctly clustered using PCA. The accuracy of the SIMCA classification for differentiation of four pomegranate groups was 82%. Also, the overall discriminant power of PLS-DA classes was about 96%, and 95% for calibration and validation sample sets, respectively. Due to the misclassification among different classes of immature pomegranates, that was lower than the expected, it was not possible to discriminate all the immature samples in individual classes. However, when considering only the two main categories of "immature" and "mature", a reasonable separation between the classes were obtained using supervised pattern recognition methods of SIMCA and PLS-DA. The SIMCA based on PCA modeling could correctly categorize the samples in two classes of immature and mature with classification accuracy of 100%.

Digital image-based tracing of geographic origin, winemaker, and grape type for red wine authentication

This work proposes the development of a simple, fast, and inexpensive methodology based on color histograms (obtained from digital images), and supervised pattern recognition techniques to classify red wines produced in the São Francisco Valley (SFV) region to trace geographic origin, winemaker, and grape variety. PCA-LDA coupled with HSI histograms correctly differentiated all of the SFV samples from the other geographic regions in the test set; SPA-LDA selecting just 10 variables in the Grayscale + HSI histogram achieved 100% accuracy in the test set when classifying three different SFV winemakers. Regarding the three grape varieties, SPA-LDA selected 15 variables in the RGB histogram to obtain the best result, misclassifying only 2 samples in the test set. Pairwise grape variety classification was also performed with only 1 misclassification. Besides following the principles of Green Chemistry, the proposed methodology is a suitable analytical tool; for tracing origins, grape type, and even (SFV) winemakers.

Extraction of artefactual MRS patterns from a large database using non-negative matrix factorization.

Despite the success of automated pattern recognition methods in problems of human brain tumor diagnostic classification, limited attention has been paid to the issue of automated data quality assessment in the field of MRS for neuro-oncology. Beyond some early attempts to address this issue, the current standard in practice is MRS quality control through human (expert-based) assessment. One aspect of automatic quality control is the problem of detecting artefacts in MRS data. Artefacts, whose variety has already been reviewed in some detail and some of which may even escape human quality control, have a negative influence in pattern recognition methods attempting to assist tumor characterization. The automatic detection of MRS artefacts should be beneficial for radiology as it guarantees more reliable tumor characterizations, as well as the development of more robust pattern recognition-based tumor classifiers and more trustable MRS data processing and analysis pipelines. Feature extraction methods have previously been used to help distinguishing between good and bad quality spectra to apply subsequent supervised pattern recognition techniques. In this study, we apply feature extraction differently and use a variant of a method for blind source separation, namely Convex Non-Negative Matrix Factorization, to unveil MRS signal sources in a completely unsupervised way. We hypothesize that, while most sources will correspond to the different tumor patterns, some of them will reflect signal artefacts. The experimental work reported in this paper, analyzing a combined short and long echo time 1 H-MRS database of more than 2000 spectra acquired at 1.5T and corresponding to different tumor types and other anomalous masses, provides a first proof of concept that points to the possible validity of this approach.

Diagnosis of pregnancy based classification of embryo culture medium samples by infrared spectrometry and chemometrics

Attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy coupled with chemometric procedures was utilized to examine embryo culture medium (ECM) samples in in vitro fertilization (IVF) to differentiate between positive and negative pregnancy.Principal component analysis (PCA) method was successfully applied to mid-IR spectra as outlier detection by application of Hotelling T2 statistic procedure. Unsupervised and supervised pattern recognition techniques, such as hierarchical agglomerative cluster algorithm (HACA), soft independent modeling of class analogy (SIMCA) as a linear supervised pattern recognition technique, binned nearest neighbors (BNN) as a local classifier and counter-propagation artificial neural networks (CP-ANN) as a non-linear technique were used in spectral data analysis for classification and differentiation. Also, the genetic algorithm (GA) as the variable reduction was employed to develop the diagnostic model.A total of 46 ECM samples were examined in the 3000–1000 cm−1 spectral region. Classification efficiency parameters including accuracy (ACC), error rates (ER), sensitivity (Sens), specificity (Spec) and non-error rate (NER) were calculated. Classification ACC of 92.00% and ER of 8.300% was acquired using GA-BNN algorithm, while GA-CP-ANN and GA-SIMCA algorithms provided 93.00% and 5.00% of ACC and ER for the test set, respectively. The results of this study presented that the GA-SIMCA and the GA-CP-ANN chemometric approaches using the infrared data, achieved acceptable results. Also, both of methods could provide the Sens rate of 100.00% and 91.00%, and the Spec rate of 91.00% and 100.00% for positive pregnancy samples (class 1) and negative ones (class 2) in prediction test, respectively.In general, the main advantages of the proposed method could be represented as a reliable, fast, simple, and accurate method for the classification and discrimination of the ECM samples based on their serum beta- subunit human chorionic gonadotropin (β-hCG) levels after embryo transfer (ET) about 15 days later.

Discrimination of vegetable oil types using Fourier transforms near infrared spectroscopy coupled with pattern recognition techniques

The aim of this research was to investigate the potential of near infrared spectroscopy (NIRs) coupled with pattern recognition techniques for discriminating of vegetable oil types (i.e. coconut oil, olive oil, rice bran oil, sesame oil, soybean oil and sun flower oil). Principle component analysis (PCA) was performed for clustering vegetable oil types. Five of supervised pattern recognition techniques such as soft independent modelling of class analogies (SIMCA), Partial least squares-discriminant analysis (PLS-DA), k-nearest neighbor (k-NN), support vector machine (SVM) and artificial neural network (ANN) were used to identify vegetable oil types. The PCA model could separate coconut oil from other vegetable oils. Two PLS-DA and SVM models showed 100% of precision, recall F-measure and accuracy for all vegetable oil whilst remainder techniques achieved a satisfactory classified performance. All supervised models could discriminate coconut oil from other oils with precision, recall F-Measure and accuracy of 100%. It seems that NIRs technique coupled with pattern recognition techniques is possible for discriminating vegetable oil types.

Assessing Geographical Origin of Gentiana Rigescens Using Untargeted Chromatographic Fingerprint, Data Fusion and Chemometrics.

Gentiana rigescens Franchet, which is famous for its bitter properties, is a traditional drug of chronic hepatitis and important raw materials for the pharmaceutical industry in China. In the study, high-performance liquid chromatography (HPLC), coupled with diode array detector (DAD) and chemometrics, were used to investigate the chemical geographical variation of G. rigescens and to classify medicinal materials, according to their grown latitudes. The chromatographic fingerprints of 280 individuals and 840 samples from rhizomes, stems, and leaves of four different latitude areas were recorded and analyzed for tracing the geographical origin of medicinal materials. At first, HPLC fingerprints of underground and aerial parts were generated while using reversed-phase liquid chromatography. After the preliminary data exploration, two supervised pattern recognition techniques, random forest (RF) and orthogonal partial least-squares discriminant analysis (OPLS-DA), were applied to the three HPLC fingerprint data sets of rhizomes, stems, and leaves, respectively. Furthermore, fingerprint data sets of aerial and underground parts were separately processed and joined while using two data fusion strategies (“low-level” and “mid-level”). The results showed that classification models that are based OPLS-DA were more efficient than RF models. The classification models using low-level data fusion method built showed considerably good recognition and prediction abilities (the accuracy is higher than 99% and sensibility, specificity, Matthews correlation coefficient, and efficiency range from 0.95 to 1.00). Low-level data fusion strategy combined with OPLS-DA could provide the best discrimination result. In summary, this study explored the latitude variation of phytochemical of G. rigescens and developed a reliable and accurate identification method for G. rigescens that were grown at different latitudes based on untargeted HPLC fingerprint, data fusion, and chemometrics. The study results are meaningful for authentication and the quality control of Chinese medicinal materials.

Identification of Cigarette Brands by Soft Independent Modeling of Class Analogy of Volatile Substances.

IntroductionThis study aimed to develop a method for discriminating cigarette brands based on the profiles of volatile components extracted from the tobacco fraction of the finished cigarettes to authenticate branded cigarettes of unknown origin.MethodsAn analytical method comprising direct thermal desorption coupled with gas chromatography-quadrupole time-of-flight mass spectrometry was developed for acquiring volatile profiles of cigarettes. About 290 samples of commercially available cigarettes were analyzed. Within this batch, 123 samples represented four popular cigarette brands. They were selected for in-depth characterization. Multivariate analysis was used to investigate the interrelations among volatile compounds of cigarettes and to identify characteristic markers for the cigarette discrimination. Supervised pattern recognition techniques were used for designing classification models.ResultsPrincipal component analysis covering all detected volatiles allowed the differentiation of cigarettes based on the brand. A number of 56 volatile components were identified as markers with high discrimination power. These compounds were used for establishing classification models. A method of soft independent modeling of class analogy developed for the four studied cigarette brands proved to be efficient in the classification of unknown cigarettes, with accuracy between 95.9% and 100%.ConclusionsThe data evaluation by soft independent modeling of class analogy was highly accurate in classification of unknown cigarettes with a low rate of false positives and false negatives. The developed models can be used for discrimination of genuine from non-genuine products with high level of probability.ImplicationsProfiling of volatiles, which is commonly used for authentication of different food commodities, was applied for the characterization of cigarette tobacco for the purpose of authentication a cigarette brand. Volatile components with a high discrimination power were identified by means of multivariate statistical methods and used for establishing of a classification model. The classification model was able to discriminate genuine from non-genuine cigarettes with a high level of prediction accuracy. This model could be a powerful tool for tobacco control to judge the authenticity of cigarettes.

FEMa: a finite element machine for fast learning

Machine learning has played an essential role in the past decades and has been in lockstep with the main advances in computer technology. Given the massive amount of data generated daily, there is a need for even faster and more effective machine learning algorithms that can provide updated models for real-time applications and on-demand tools. This paper presents FEMa—a finite element machine classifier—for supervised learning problems, where each training sample is the center of a basis function, and the whole training set is modeled as a probabilistic manifold for classification purposes. FEMa has its theoretical basis in the finite element method, which is widely used for numeral analysis in engineering problems. It is shown FEMa is parameterless and has a quadratic complexity for both training and classification phases when basis functions are used that satisfy certain properties. The proposed classifier yields very competitive results when compared to some state-of-the-art supervised pattern recognition techniques.

Classification of saline water for irrigated agriculture using near infrared spectroscopy coupled with pattern recognition techniques

This research aimed to create near infrared (NIR) spectroscopy models for the classification of saline water with a pattern recognition technique. A total of 112 water samples were collected from the Tha Chin river basin in Thailand. Water samples with salinity less than 0.2 g/l were identified as suitable for agriculture, while water samples with salinity higher than 0.2 g/l were found to be unsuitable. The NIR spectra of water samples were recorded using a Fourier transform (FT) NIR spectrometer in the wavenumber of 12,500–4,000 cm-1. The salinity of each water sample was analysed by electrical conductivity meter. Identification models were established with 5 supervised pattern recognition techniques including k-nearest neighbour (k-NN), support vector machine (SVM), artificial neural network (ANN), soft independent modelling of class analogies (SIMCA), and partial least squares-discriminant analysis (PLS-DA). The performance of the NIR model was carried out with a split-test method. About 80% of spectra (90 spectra) were randomly selected to develop the classification models. After model development, the NIR spectroscopy models were used to classify the categories of the remaining samples (22 samples). The ANN model showed the highest performance for classifying saline water with precision, recall, F-measure and accuracy of 84.6%, 100.0%, 91.7% and 90.9%, respectively. Other techniques presented satisfactory classification results with accuracy greater than 68.2%. This point indicated that NIR spectroscopy coupled with the pattern recognition technique could be applied to classify saline water for agricultural use according to salinity level in natural resources.

Tracing the Geographical Origin of Lentils (Lens culinaris Medik.) by Infrared Spectroscopy and Chemometrics

The feasibility of applying the infrared spectroscopy for the geographical origin traceability of lentils from two different countries (Italy and Canada) was investigated. In particular, lentil samples were analyzed by Fourier transform near- and mid-infrared (FT-NIR and FT-MIR) spectroscopy and then discriminated by applying supervised models, i.e., linear discriminant analysis (LDA) and partial least squares discriminant analysis (PLS-DA). To avoid LDA overfitting, two variable strategies were adopted, i.e., a variable reduction by principal component analysis and a variable compression by wavelet packet transform algorithm. FT-MIR models were more discriminating compared to FT-NIR ones with prediction abilities ranging from 98 to 100% and from 91 to 100% for cross- and external validation, respectively. The combination of the FT-MIR and FT-NIR data did not improve the model performances. These findings demonstrated the suitability of the FT-MIR spectroscopy, in combination with supervised pattern recognition techniques, to successfully classify lentils according to their geographical origin.

GENETIC MARKERS AND BIOSTATISTICAL METHODS AS APPROPRIATE TOOLS TO PRESERVE GENETIC RESOURCES

The aim of presented study was to assess the most suitable way how to distinguishdifferent breeds based on molecular markers. One of the most difficult aspects ofquality assurance schemes is their reliability. The verification of fraud needs greatefforts in control strategies. The use of DNA markers has been shown to be auseful tool for individual identification. It is necessary to use modern statisticalmethod based on data mining and supervised learning. Supervised patternrecognition techniques use the information about the class membership of thesamples to a certain group (class or category) in order to classify new unknownsamples in one of the known classes on the basis of its pattern of measurements.Large scale of supervised learning oriented method was used for traceability andidentification on individual level. A result of provided study shows the possibilityto classify unknown samples according to genetic data. Model is also useful forclassification on many logical levels as brand, region and many others. If we takein the account only Slovak and Austrian Pinzgau cattle, based on SNP chip data, itis not possible to separate them using Bayesian approach. Once we considered withthe admixture of breeds involved in the historical development as well asinbreeding, selection signatures and migration, we were able to separate evengenetically similar breeds. It is possible distinguish between closely relatedpopulations based on different markers. We just need to select the appropriate typeof analysis.

Classification of cowpea beans using multielemental fingerprinting combined with supervised learning

Multielemental compositions (Ag, As, Ba, Be, Cd, Cs, Co, Cr, Cu, Mo, Ni, Pb, Sb, Se, Sn, Sr, Tl, Rb, V, and Zn) of 106 cowpea bean samples belonging to different varieties collected from the province of Corrientes in Argentina were determined using inductively coupled plasma mass spectrometry (ICP-MS). Based on the multielemental data, five supervised learning techniques, namely, linear discriminant analysis (LDA), partial least square discriminant analysis (PLS-DA), k nearest neighbors (k-NN), random forest (RF), and support vector machine (SVM) with radial basis function Kernel, were computed aiming at building classification models that allow one to predict the botanical variety of the samples based on their element profiles. The best classification performance was obtained by SVM with 93% accuracy rate. The model developed through this method enabled the correct separation of the samples into the five cowpea varieties investigated, where 100% sensitivity was achieved for most of the predicted classes. Thus, SVM was the algorithm selected for the classification of the cowpea beans according to their botanical variety. Multielemental determination coupled with supervised pattern recognition techniques have proved to be an interesting approach for differentiating a diverse range of cowpea genotypes. This study has contributed toward generalizing the use of multielemental fingerprinting as a promising tool for testing the authenticity of cowpea beans on a global scale.

Chemometrics advances on the challenges of the gas chromatography–mass spectrometry metabolomics data: a review

Metabolomics refers to the comprehensive and quantitative analysis of biological small molecules and aims to gather as much metabolic information as possible from a biological system. Accurate quantification of all metabolites in complex samples is a challenging issue and requires mathematical methods to analyze complex and very big data. Chemometrics methods have become crucial and dedicated tools for extracting valuable information from complex data. Additionally, gas chromatography–mass spectrometry (GC–MS) is a key tool with great potential in many analytical fields and has been demonstrated to be capable of facing many important challenges related to metabolomics research. This review presents an overview of the major advances in data pre-processing methods, metabolite identification and statistical data analysis methods for analyzing metabolomics data collected using GC–MS tool. Moreover, current study provides new insights into important chemometrics methods such as baseline correction, noise reduction, alignment, multivariate curve resolution, metabolite identification, multi-way calibration, unsupervised and supervised pattern recognition techniques to address metabolomics problems. For the sake of clarity, each of these topics will be discussed with different examples from the literature.

Quantitative and qualitative analysis of polycyclic aromatic hydrocarbons in urine samples using a non-separative method based on mass spectrometry

In this work, a method for the quantitative and qualitative analysis of 11 polycyclic aromatic hydrocarbons (PAHs) in urine samples is reported. The method is based on the coupling of a programmed temperature vaporizer (PTV) with a quadrupole mass spectrometer (qMS), via a deactivated fused silica tubing. Before the PTV-qMS analysis, the samples were subjected to a liquid-liquid extraction (LLE).The method was rapid since no chromatographic separation was performed. The samples were introduced directly into the PTV, and the analytes were trapped in the Tenax-TA® packed liner while the solvent was purged. After that, all the compounds reached the mass spectrometer, obtaining the fingerprint of the analysed samples.Urine samples free of PAHs and the same samples spiked with the compounds were analysed. The resulting profile signals were used to quantify the analytes using multivariate calibration, and to classify the samples according to the presence or absence of the PAHs. In the latter task, non-supervised and supervised pattern recognition techniques were employed. The calibration models worked satisfactorily and errors lower or equal to 15% were obtained, in most cases, when an external validation set was analysed. Regarding the classification of the samples, most of the supervised pattern recognition techniques provided excellent results (100% success), where all of the samples were classified correctly.