Chemometrics and Intelligent Laboratory Systems最新文献

{"title":"A sensitive efficient multiple predictable feature extraction and fusion method for complicated industrial fault detection","authors":"Xiaogang Deng,&nbsp;Yujiang Wang,&nbsp;Wenjie Yang","doi":"10.1016/j.chemolab.2025.105423","DOIUrl":"10.1016/j.chemolab.2025.105423","url":null,"abstract":"<div><div>In recent years, graph-based predictable feature analysis (GPFA) has emerged as a potential tool in industrial fault diagnosis. However, the basic GPFA can only extract the linear predictable feature information, which is incompetent in the scenarios of the complicated linear-nonlinear hybrid data characteristic. To handle this issue, an improved fault detection method, called Sensitive Efficient Multiple Graph-based Predictable Feature Analysis (SEM-GPFA), is presented for mining the linear and nonlinear predictable features simultaneously. In this method, a multiple predictable feature fusion framework is constructed to utilize the complementary advantages of linear and nonlinear features fully. The linear predictable features are extracted by the basic GPFA, while the nonlinear predictable features are captured by the nonlinear GPFA model. To address the high computational complexity of traditional kernel methods, a random Fourier mapping method is used to improve the nonlinear feature extraction approach, enhancing operational efficiency. Considering that the specific fault information may be concealed in massive features of the model, a fault-sensitive feature highlighting strategy is designed by assigning relatively large weights to emphasize the influence of significant fault features. Finally, case studies on the Continuously Stirred Tank Reactor (CSTR) process and the Tennessee Eastman (TE) chemical system are conducted to demonstrate the superiority of the proposed method.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105423"},"PeriodicalIF":3.7,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143922230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computing eccentricity based topological indices of silicate network with applications to QSPR/QSAR analysis","authors":"Deepika S.,&nbsp;Arathi P.","doi":"10.1016/j.chemolab.2025.105424","DOIUrl":"10.1016/j.chemolab.2025.105424","url":null,"abstract":"<div><div>Topological indices are one of the useful tools in graph theory provided by chemists, which is a numeric quantity that helps to predict physico-chemical properties of chemical compounds. To classify molecules and model unknown structures the topological representations of molecular structures with the necessary properties can be used. The topological indices of distance-based methods are advanced tools that show effective significance in chemical graph theory (CGT). In recent years, many topological indices (TIs) have been studied and applied in theoretical chemistry and have been widely used to investigate quantitative structure-property relationships (QSPR), and quantitative structure-activity relationship (QSAR) research, and to evaluate networks. In this paper, the various eccentricity-based topological indices were calculated for the silicate network and plotted as 3D graphs for each calculated indices. In contrast, physico-chemical properties such as ‘boiling point (BP), enthalpy (E), molecular weight (MW), complexity (C), molar refractivity (MR), polar surface area (PSA), and refractive index (RI)’ were computed for silicate compounds using a linear regression model and plotted graphs for each properties with the values of correlated coefficients. These results have been used for the development of drug delivery systems.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105424"},"PeriodicalIF":3.7,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143916804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing soil texture classification accuracy based on VNIR lab spectroscopy","authors":"Ternikar Chirag Rajendra ,&nbsp;Cécile Gomez ,&nbsp;Subramanian Dharumarajan ,&nbsp;D. Nagesh Kumar","doi":"10.1016/j.chemolab.2025.105419","DOIUrl":"10.1016/j.chemolab.2025.105419","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Soil texture is an important soil parameter controlling various physical, chemical and biological soil properties. Visible Near-Infrared (VNIR) spectroscopy has garnered attention due to its simplicity, non-destructive nature, absence of hazards and rapidity. Due to its popularity, numerous studies employ this technique without adhering to the unity constraint on predicted fractions. This study aims to assess the accuracy of soil texture classification in the USDA textural triangle through laboratory VNIR spectra. Five different approaches were evaluated in this study: i) four approaches (&lt;em&gt;A1-A4&lt;/em&gt;), defined as regression-assisted classification techniques, were based on the Partial Least Squares Regression (PLSR) method to predict quantitative fractions followed by a texture classification based on the USDA texture triangle and ii) one approach (&lt;em&gt;A5&lt;/em&gt;), defined as a direct classification method, was based on the Partial Least Squares Discriminant Analysis (PLS-DA) classifier to classify soil texture using spectra directly. Each regression-assisted classification approach varies in predicting fractions and ensuring the unity constraint on the predicted fractions. In approach &lt;em&gt;A1&lt;/em&gt;, the clay, silt and sand fractions predicted by PLSR for each sample were normalized to ensure sum-to-unity. In approach &lt;em&gt;A2&lt;/em&gt;, the silt content was derived as residual from the clay and sand contents predicted by PLSR for each sample, ensuring unity. In Approach &lt;em&gt;A3&lt;/em&gt;, the clay, silt and sand fractions were simultaneously predicted using a multi-output variant of PLSR. Approach &lt;em&gt;A4&lt;/em&gt; employed PLSR on log-ratio transformed (LRT) fractions, enabling simultaneous prediction and inherently ensuring sum-to-unity. Approach &lt;em&gt;A4&lt;/em&gt; via LRT utilizes information about the relative fractions of soil texture instead of the absolute fractions. For the regression-based fraction predictions, approaches (&lt;em&gt;A1-A4&lt;/em&gt;) achieved similar performances, with mean coefficients of determination (R&lt;sup&gt;2&lt;/sup&gt;) of 0.88–0.90 for clay (RMSE: 4.2–4.4 %), 0.82–0.84 for sand (RMSE: 6.1–6.5 %), but lower (R&lt;sup&gt;2&lt;/sup&gt; = 0.29–0.38) for silt (RMSE: 3.8–4.1 %). Approach &lt;em&gt;A2&lt;/em&gt;, which infers silt as a residual, yielded poorer silt predictions. Despite these quantitative differences, the resulting classification accuracies in the USDA texture triangle were high with overall accuracy of 71–71.8 %, average accuracy of 62.4–65.3 % and Cohen's Kappa of 0.61–0.62 for &lt;em&gt;A1&lt;/em&gt;-&lt;em&gt;A4&lt;/em&gt; while &lt;em&gt;A5&lt;/em&gt;, attained only 56.4 % overall accuracy and Cohen's Kappa of 0.42. Among the regression-assisted methods, Approach &lt;em&gt;A4&lt;/em&gt; using log-ratio transformations of clay, silt, and sand simultaneously enforced compositional constraints and matched the best classification performances (OA = 71.4 %, AA = 65.3 %, K = 0.62) while requiring fewer models. This work highlighted that i) the four regression-assisted classification approaches provided comparabl","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105419"},"PeriodicalIF":3.7,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identification of Monteverdia ilicifolia by fourier-transform mid-infrared spectroscopy associated with chemometrics and machine learning","authors":"Ahmad Kassem El Zein ,&nbsp;Alexandre de Fátima Cobre ,&nbsp;Raul Edison Luna Lazo ,&nbsp;Kevin Alves Antunes ,&nbsp;Jane Manfron ,&nbsp;Luana Mota Ferreira ,&nbsp;Roberto Pontarolo","doi":"10.1016/j.chemolab.2025.105420","DOIUrl":"10.1016/j.chemolab.2025.105420","url":null,"abstract":"<div><div><em>Monteverdia ilicifolia</em> (Mart. Ex Reissek) Biral, a member of the Celastraceae botanical family, is widely recognized for its broad-spectrum therapeutic effects in South America, particularly in Brazil, where it is commonly referred as “espinheira-santa”. This study aimed to develop a chemometric and machine learning-based method for to accurately identify and differentiate <em>M. ilicifolia</em> from morphologically similar species used as adulterants. Fourier transform mid-infrared spectrometry (MIR-FTIR) was used to analyze leaves (n = 6 species, 3000 spectra), powders (n = 6 species, 3000 spectra) and extracts samples (n = 6 species, 600 spectra). The spectral datasets were predicted by Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA). The PLS-DA model was challenged with samples of other common plant species (n = 3) and commercial available <em>M. ilicifolia</em> (n = 10) to evaluate its predictive capability. PCA successfully distinguished between the plant species. PLS-DA achieved superior performance with extract samples, exhibiting sensitivity, specificity and accuracy of 94, 100 and 99 %, respectively. Machine learning algorithms were developed to better represent the leaves and powder samples through Random Forest and 10-fold validation methodology. The model yielded high accuracy in all sample types, with low false positive rate and excellent performance across the metrics of accuracy, recall, precision, F1 Score, Kappa index and Matthews Correlation Coefficient (MCC). PCA and PLS-DA models presented limitations over the complexity of leaves and powders samples. Machine learning algorithms showed robustness and flexibility, proving to be effective in the detection and discrimination of <em>M. ilicifolia</em>.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105420"},"PeriodicalIF":3.7,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143922229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative analysis of continuous similarity measures for compound identification in mass spectrometry-based metabolomics","authors":"Hunter Dlugas ,&nbsp;Xiang Zhang ,&nbsp;Seongho Kim","doi":"10.1016/j.chemolab.2025.105417","DOIUrl":"10.1016/j.chemolab.2025.105417","url":null,"abstract":"<div><div>In mass spectrometry (MS)-based metabolomics, the most straightforward and efficient approach for compound identification is the comparison of similarity scores between experimental spectra and reference spectra. Among various single and composite similarity measures, the Cosine Correlation is favored due to its simplicity, efficiency, and effectiveness. Recently, the Shannon Entropy Correlation has shown superior performance over several other measures, including the Cosine Correlation, in LC-MS-based metabolomics, particularly concerning receiver operating characteristic (ROC) curves and false discovery rates. However, previous comparisons did not consider the weight factor transformation, which is critical for achieving higher accuracy with the cosine correlation. This study conducted a comparative analysis of the Cosine Correlation and Shannon Entropy Correlation, incorporating the weight factor transformation during preprocessing. Additionally, we developed a novel entropy correlation measure, the Tsallis Entropy Correlation, which offers greater versatility than the Shannon Entropy Correlation. Our accuracy-based results indicate that the weight factor transformation is essential for achieving higher identification performance in both LC-MS and GC-MS-based compound identification. Although the Tsallis Entropy Correlation outperforms the Shannon Entropy Correlation in terms of accuracy, it comes with higher computational expense. In contrast, the Cosine Correlation, when combined with the weight factor transformation, achieves the highest accuracy and the lowest computational expense, demonstrating both robustness and efficiency in MS-based compound identification.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105417"},"PeriodicalIF":3.7,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143916803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Functionality of rotational ambiguity in self-modeling methods to signal contribution of chemical components","authors":"Hamideh Bakhshi ,&nbsp;Hamid Abdollahi ,&nbsp;Róbert Rajkó","doi":"10.1016/j.chemolab.2025.105410","DOIUrl":"10.1016/j.chemolab.2025.105410","url":null,"abstract":"<div><div>Self-modeling curve resolution (SMCR) methods are widely acknowledged as potent tool in chemometrics, facilitating the decomposition of bilinear data matrices into chemically interpretable matrices. Nonetheless, these methods frequently yield results with ambiguities, particularly rotational ambiguity, leading to non-uniqueness of outcomes. This study investigates the influence of signal contributions of chemical components on rotational ambiguity. Utilizing simulated data from HPLC-DAD and one real excitation emission fluorescence data, the impacts of signal contributions of chemical components on spectral and concentration profiles are assessed. The findings illustrate that increasing the signal contribution of a chemical component can mitigate rotational ambiguity. Furthermore, the efficacy of employing second-order standard addition in reducing rotational ambiguity and enhancing the accuracy of quantitative analyses is examined.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"262 ","pages":"Article 105410"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Significance determination of individual metabolic abundance changes owing to environmental impacts: Factorial design t-distribution spectral representations","authors":"Gustavo G. Marcheafave ,&nbsp;Leonardo J. Duarte ,&nbsp;Elis D. Pauli ,&nbsp;Ieda S. Scarminio ,&nbsp;Roy E. Bruns","doi":"10.1016/j.chemolab.2025.105413","DOIUrl":"10.1016/j.chemolab.2025.105413","url":null,"abstract":"<div><div>Currently quantitative metabolic analysis is work-intensive and time-consuming normally demanding the use of both chromatography and mass spectrometry. Screening spectral data with principal component analysis is fast and helps identify metabolites. However, it has not been used for quantitative analysis as it cannot determine the statistical significance of individual metabolite abundance changes owing to simulated environmental impacts. Factorial design t univariate spectral representations provide a relatively fast and simple method to determine the statistical significance of individual NMR channels forming peaks and help fill the gap between qualitative and quantitative metabolic analysis of plants suffering environmental impacts. These composite spectral representations, introduced and described for the first time, are univariate statistical t values calculated from factorial design spectra plotted as a function of the analytical channels. They are simple to understand by chemists and biologists with limited statistical knowledge as they only use one basic statistical t distribution equation. We demonstrate their usefulness with factorial design analyses of ¹H NMR spectra of yerba mate samples obtained from different solvent extracts of a mixture design. The spectral representation peak locations are almost the same as those of principal component loadings of ASCA effect matrices although their peak heights are much different and correspond to the statistical significance levels of individual metabolic abundance changes. Spectral representations have ordinates of calculated t values and results for different data sets can be analyzed simultaneously on a common graph whereas this is not possible for loadings that correspond to different PCA coordinate spaces.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105413"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Supervised laser induced breakdown spectroscopy classification for prehistoric chert provenance: A methodological framework","authors":"Julien Le Guirriec ,&nbsp;Jonàs Alcaina-Mateos ,&nbsp;Bruno Bousquet ,&nbsp;François-Xavier Le Bourdonnec ,&nbsp;Marta Sánchez de la Torre","doi":"10.1016/j.chemolab.2025.105411","DOIUrl":"10.1016/j.chemolab.2025.105411","url":null,"abstract":"<div><div>The sourcing of lithic artefacts, and in particular chert, is a key proxy for archaeologist to better understand hunter-gatherer populations. The heterogeneous structure of chert samples, low trace element content, and variability of outcrops make the geochemical characterisation a methodological challenge. Laser Induced Breakdown Spectroscopy (LIBS) can overcome the issues of variability by enabling fast and low-cost and nearly non-destructive analysis of samples. This paper presents a protocol for chert sourcing using LIBS, by comparing several supervised classification models and signal pre-processing with two different approaches: a selection of features or the use of broadband spectra, with an example application on marine flysch cherts from the central Pyrenes (France) presenting similar micropaleontological and textural features.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105411"},"PeriodicalIF":3.7,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced data point importance for subset selection in partial least squares regression: A comparative study with Kennard-Stone method","authors":"Mahia V. Solout ,&nbsp;Somaye Vali Zade ,&nbsp;Hamid Abdollahi ,&nbsp;Jahan B. Ghasemi","doi":"10.1016/j.chemolab.2025.105416","DOIUrl":"10.1016/j.chemolab.2025.105416","url":null,"abstract":"<div><div>In multivariate data analysis, the selection of representative subsets of samples is crucial for developing accurate predictive models. This study evaluates the application of the Enhanced Data Point Importance (EDPI) method for subset selection, comparing its performance with the widely-used Kennard-Stone algorithm. The EDPI method ranks all the data points using the DPI and layered convex hull approach, resulting in a ranked sequence of points based on their importance in the dataset, with the most important point being the most informative. Both methods were applied to two distinct datasets, and Partial Least Squares Regression (PLSR) models were developed for each subset to assess predictive performance. The EDPI method demonstrated comparable performance to the Kennard-Stone method across various sample sizes. The EDPI-PLS models achieved lower Root Mean Square Error of Prediction (RMSEP) values with fewer samples, indicating efficient subset selection, and the method is less inclined to select the influential points in the dataset. Moreover, the running time analysis highlighted the computational efficiency of the EDPI method, especially in high-dimensional datasets. These findings suggest that EDPI is a robust and informative strategy for sample subset selection, offering advantages in predictive accuracy and computational efficiency.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105416"},"PeriodicalIF":3.7,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI in MRI brain tumor diagnosis: A systematic review of machine learning and deep learning advances (2010–2025)","authors":"Vaidehi Satushe ,&nbsp;Vibha Vyas ,&nbsp;Shilpa Metkar ,&nbsp;Davinder Paul Singh","doi":"10.1016/j.chemolab.2025.105414","DOIUrl":"10.1016/j.chemolab.2025.105414","url":null,"abstract":"<div><div>Brain tumors present critical health challenges due to abnormal tissue growth within the brain, potentially leading to life-threatening conditions if left untreated. MRI stands as the primary diagnostic tool for identifying brain tumors, offering superior resolution and tissue differentiation compared to other imaging modalities. This systematic literature review explores the application of ML and DL techniques in enhancing the diagnosis of brain tumors from MRI scans. ML and DL algorithms, particularly CNNs, have demonstrated significant success in automating the detection and classification of brain tumors by analyzing complex imaging patterns. The review follows PRISMA guidelines, synthesizing findings from studies between 2010 and 2025. Key themes include the utilization of diverse datasets, advanced feature extraction methods, and the computational efficiency of ML and DL models. Despite notable advancements, challenges such as data diversity and model interpretability persist, underscoring the need for ongoing research to optimize these techniques for enhanced clinical outcomes in brain tumor diagnosis. The review discusses the effectiveness of CNNs, SVMs, ensemble methods, transformers and other ML approaches in improving diagnostic accuracy and reliability. It also addresses future research directions aimed at overcoming current limitations and advances in the field.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"263 ","pages":"Article 105414"},"PeriodicalIF":3.7,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143902299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}