Chemometrics and Intelligent Laboratory Systems最新文献

{"title":"Enhanced satellite image resolution with a residual network and correlation filter","authors":"Ajay Sharma ,&nbsp;Bhavana P. Shrivastava ,&nbsp;Praveen Kumar Tyagi ,&nbsp;Ebtasam Ahmad Siddiqui ,&nbsp;Rahul Prasad ,&nbsp;Swati Gautam ,&nbsp;Pranshu Pranjal","doi":"10.1016/j.chemolab.2024.105277","DOIUrl":"10.1016/j.chemolab.2024.105277","url":null,"abstract":"<div><div>This study addresses the predominant challenge of very low-resolution satellite images in remote sensing applications, a common issue in satellite image-based surveillance. Existing satellite image recognition algorithms struggle with such low-resolution images, and traditional Super-Resolution (SR) techniques fall short for very low-resolution cases. We propose the Progressive Satellite Image Super-Resolution (PSISR) model to bridge this gap. Unlike current learning-based SR methods, the PSISR model specifically targets very low-resolution satellite images. In satellite image super-resolution, problems with feature fusion that result in image noise, blind spots, poor perceptual quality, and checkboard artifacts are encountered during the reconstruction process. Current models try to improve perceptual quality, but they frequently show challenges in attaining acceptable outcomes because of losses during reconstruction. Using a combined loss function, correlation filters, and a loss-aware upscaling network layer, the PSISR model presents a revolutionary methodology. The model adopts a cascading structure with dense skip connections, sequentially upscaling images by factors of <span><math><mrow><mn>2</mn><mo>×</mo></mrow></math></span>, <span><math><mrow><mn>4</mn><mo>×</mo></mrow></math></span>, and <span><math><mrow><mn>8</mn><mo>×</mo></mrow></math></span> through three modules. To validate the model's superiority, a study is conducted, confirming its effectiveness compared to baseline models and also training the other models using the available dataset to prove the effectiveness of the model. The PSISR model effectively addresses the challenge of extracting more features with minimal losses, resulting in high magnification during reconstruction. Our method outperforms state-of-the-art techniques, including Swin2-MoSE, MambaFormer, SRFBN and RCAN, with a PSNR improvement of up to 0.4 dB and a 0.003 SSIM enhancement across various datasets. This demonstrates the effectiveness of our approach in producing high-quality outputs, achieving a 99.25 % correlation efficiency between the generated and original images.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"256 ","pages":"Article 105277"},"PeriodicalIF":3.7,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142745557","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":"A novel soft sensor approach for industrial quality prediction based TCN with spatial and temporal attention","authors":"Lei Zhang ,&nbsp;Guofeng Ren ,&nbsp;Shanlian Li ,&nbsp;Jinsong Du ,&nbsp;Dayong Xu ,&nbsp;Yinhua Li","doi":"10.1016/j.chemolab.2024.105272","DOIUrl":"10.1016/j.chemolab.2024.105272","url":null,"abstract":"<div><div>The complex industrial process is often characterized by strong multivariate coupling and nonlinear dynamic changes, which pose great challenges to modeling and prediction. Traditional deep learning methods are difficult to effectively capture spatiotemporal characteristics of industrial processes, resulting in poor prediction accuracy. To tackle this issue, we propose a novel end-to-end method named STA-TCN, which utilizes a temporal convolutional network (TCN) with both spatial and temporal attention mechanisms. The TCN uses causal and dilated convolutions to capture long temporal patterns in time series data. The spatial attention identifies the significance of different features, while the temporal attention focuses on crucial time steps. This design assigns adaptive weights to different features and emphasizes key moments to improve the accuracy of dynamic processes. We conduct experiments on two industrial datasets and show that the proposed STA-TCN method achieves significantly improved predictive performance compared to TCN for quality prediction of industrial processes. The results validate the effectiveness and robustness of the proposed method.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"257 ","pages":"Article 105272"},"PeriodicalIF":3.7,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143156194","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":"GATNM: Graph with Attention Neural Network Model for Mycobacterial Cell Wall Permeability of Drugs and Drug-like Compounds","authors":"Agung Surya Wibowo ,&nbsp;Osphanie Mentari Primadianti ,&nbsp;Hilal Tayara ,&nbsp;Kil To Chong","doi":"10.1016/j.chemolab.2024.105265","DOIUrl":"10.1016/j.chemolab.2024.105265","url":null,"abstract":"<div><div><em>Mycobacterium tuberculosis</em> cell wall has complexity and unusual organization. These conditions make the nutrients and antibiotics difficult to penetrate this wall which affects the low activity of several antimycobacterial drugs in mycobacteria cells. Based on this information, the cell wall permeability prediction in some compounds becomes important and would help develop novel antitubercular drugs. Recently, there have been many predictions helped by computational technology using the Simplified Molecular Input Line Entry System (SMILES) input drug compounds. In this study, we applied computational technology to predict the permeability of cell walls to some compounds or drugs. We evaluated several common machine learning models for their ability to predict cell wall permeability. However, none of these models achieved satisfactory performance. We investigated a Graph with Attention Neural Network (GATNN) model to address this challenge. In the case of permeability detection, to the best of our knowledge, the GATNN model is considered a new approach to improve the prediction performance of the penetration ability of some compounds to the cell wall of the mycobacterial. Additionally, we optimized the accuracy value to get the best hyperparameter and the best model by Optuna. After getting the optimal model, by using the benchmark dataset, this model has slightly increased the performance over the previous model in accuracy and specificity to 78.9% and 81.5%. As a complementary, we also provided an ensemble model and generated the interpretability of the model. The code and materials of all experiments in this paper can be accessed freely at this link: <span><span>https://github.com/asw1982/MTbPrediction</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"256 ","pages":"Article 105265"},"PeriodicalIF":3.7,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142719739","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":"A review of quantitative structure-activity relationship: The development and current status of data sets, molecular descriptors and mathematical models","authors":"Jianmin Li ,&nbsp;Tian Zhao ,&nbsp;Qin Yang ,&nbsp;Shijie Du ,&nbsp;Lu Xu","doi":"10.1016/j.chemolab.2024.105278","DOIUrl":"10.1016/j.chemolab.2024.105278","url":null,"abstract":"<div><div>Developing Quantitative Structure-Activity Relationship (QSAR) models applicable to general molecules is of great significance for molecular design in many disciplines. This paper reviews the development and current status of molecular QSAR research, including datasets, molecular descriptors, and mathematical models. A representative bibliometric analysis reveals the evolutionary trends in this field in the past decade. Based on the discussion of the advantages and shortcomings of existing methods, the requirements and possible approaches for developing a widely applicable QSAR model were put forward. This goal poses a series of challenges to QSAR, including: (1) Having a sufficient number of structure-activity relationship instances as training data to cope with the complexity and diversity of molecular structures and action mechanisms; (2) Developing and using precise molecular descriptors to avoid the situation of ‘garbage in, garbage out’, while balancing descriptor dimensions and computational costs; and (3) Using powerful and flexible mathematical models, such as deep learning models, to learn complex functional relationships between descriptors and activity. With the emergence of larger and higher-quality data sets, more accurate molecular descriptors and deep learning methods, predictive ability, interpretability and application domain of QSAR models will continue to improve, and it will play a more important role in various fields of molecular design.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"256 ","pages":"Article 105278"},"PeriodicalIF":3.7,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142719740","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":"VAE-SIMCA — Data-driven method for building one class classifiers with variational autoencoders","authors":"Akam Petersen,&nbsp;Sergey Kucheryavskiy","doi":"10.1016/j.chemolab.2024.105276","DOIUrl":"10.1016/j.chemolab.2024.105276","url":null,"abstract":"<div><div>The paper proposes a new method for building one class classifiers based on variational autoencoders (VAE). The classification decision is built on a linear combination of two squared distances: computed for the original and the reconstructed image as well as for the representation of the original image inside the latent space formed by VAE. Because both distances are well approximated by scaled chi-square distribution, the decision boundary is computed using the theoretical quantile function for this distribution and the predefined probability for Type I error, ⍺. Thereby the boundary does not require any specific optimization and is solely based on the model outcomes computed for the training set.</div><div>The original idea of the proposed method is inherited from another OCC approach, Data Driven Soft Independent Method for Class Analogies, where singular value decomposition is employed for building the latent space. In this paper we show how this idea can be adopted to be used with VAE for detection of anomalies on images. The paper describes the theoretical background, introduces the main outcomes as well as tools for visual exploration of the classification results, and shows how the method works on several simulated and real datasets.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"256 ","pages":"Article 105276"},"PeriodicalIF":3.7,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142719850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Studies on QSAR models for the anti-virus effect of oseltamivir derivatives targeting H5N1 based on Mix-Kernel support vector machine","authors":"Xianquan Wang ,&nbsp;Fengjing Nie ,&nbsp;Zhan Gao ,&nbsp;Guoliang Li ,&nbsp;Dengshuai Zhang ,&nbsp;Jinfeng Zhang ,&nbsp;Peijian Zhang","doi":"10.1016/j.chemolab.2024.105273","DOIUrl":"10.1016/j.chemolab.2024.105273","url":null,"abstract":"<div><div>To assist in design and synthesis of novel oseltamivir derivatives as to help yield more inhibitors with lower time and cost, four quantitative structure-activity relationship (QSAR) models on 83 ODs with methods of random forest (RF), extreme gradient boosting (XGBoost), radial basis kernel function support vector machine (RBF-SVM) and mix kernel function support vector machine (MIX-SVM) were established. In the study, five descriptors of the highest importance were identified by RF. To further examine the robustness and stability of the four models, leave-one-out cross validation was adapted to the models and R, Rwere used to measure the performance. The model built by MIX-SVM performed best on the dataset: R² on training set and test set were 0.963 and 0.961, mean squared error (MSE) on training set and test set were 0.055 and 0.054, and R, R were 0.918 and 0.889, respectively. Furthermore, five important descriptors were selected for analysis, and the modelling method with the application of MIX-SVM achieved great prediction performance, which could facilitate the design and synthesis of oseltamivir derivatives.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"261 ","pages":"Article 105273"},"PeriodicalIF":3.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761165","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":"Sound uncertainty-based strategy for oil spill source identification","authors":"Ana Catarina Rocha ,&nbsp;Carla Palma ,&nbsp;Ricardo J.N. Bettencourt da Silva","doi":"10.1016/j.chemolab.2024.105275","DOIUrl":"10.1016/j.chemolab.2024.105275","url":null,"abstract":"<div><div>Oil spills are frequent and a major environmental threat, whether they are massive or small. Therefore, authorities and experts have developed analytical chemistry tools to identify spill sources and address these illegal acts by comparing oil patterns obtained by Gas Chromatography-Mass Spectrometry analysis of the spill (Sp) and suspected sources (SS) samples. Several methodologies have proposed different balances between data processing complexity and reliability. Supported by the accessibility and validity of Microsoft Excel spreadsheets, an alternative, accurate, and user-friendly tool was developed for spill source identification based on Monte Carlo Method (MCM) simulation of correlated oil components expressed by abundance ratio (<em>DR</em>). However, the statistical control of various <em>DR</em> and the degree of similarity of samples' compositions, at defined confidence levels, impact the probability of true and false composition equivalence claim of Sp and SS becoming a challenge to recognise the offender. This work not only compares the MCM and the conventional approaches allowing to highlight the limitations that result in evidence with greater uncertainty, but also offers a statistically sound strategy that manages the probabilities of a compositional equivalence claim assessing the ability to distinguish competing spill sources and reporting the most likely polluting source with reduced uncertainty. A decision chart proposed, based on objective and statistically sound criteria, indicates the performance of consecutive <em>DR</em> comparison trials if necessary. The target values established for the probability of compositional equivalence claim of the Sp and the first and second most likely SS (≥95.0 % and ≤0.50 %, respectively) provide to forensic experts’ sound evidence to be presented in court (likelihood ratio ≥190). This work represents a significant breakthrough in comparing complex chemical oil patterns.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"256 ","pages":"Article 105275"},"PeriodicalIF":3.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142719851","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":"A flame image soft sensor for oxygen content prediction based on denoising diffusion probabilistic model","authors":"Yi Liu ,&nbsp;Angpeng Liu ,&nbsp;Shuang Gao","doi":"10.1016/j.chemolab.2024.105269","DOIUrl":"10.1016/j.chemolab.2024.105269","url":null,"abstract":"<div><div>High-precision oxygen content measurement is crucial for statistical analysis of combustion chemical reaction. Deep learning based soft sensor is a new class of intelligent tools for monitoring combustion oxygen content. But in the actual production, data for sensors are often insufficient. A new soft sensing model is proposed to display the excellent performance of denoising diffusion probabilistic model (DDPM) in data generation. Firstly, a UNet based soft sensor is designed by integrating self-attention mechanism into the convolution layers. Then, a denoising loss function is designed to link the feature extraction process of soft sensor model with the reverse denoising process of DDPM, and the noise prediction neural network of DDPM is used to improve the feature extractability of the soft sensor model. Finally, the proposed model is compared with common models. The effectiveness and superiority of the proposed soft sensing model for oxygen content prediction, especially in the case with a small sample size, are both confirmed by the results.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"255 ","pages":"Article 105269"},"PeriodicalIF":3.7,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652924","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":"Prediction of potential antitumor components in Ganoderma lucidum: A combined approach using machine learning and molecular docking","authors":"Qi Yang ,&nbsp;Lihao Yao ,&nbsp;Fang Jia ,&nbsp;Guiyuan Pang ,&nbsp;Meiyu Huang ,&nbsp;Chengxiang Liu ,&nbsp;Hua Luo ,&nbsp;Lili Fan","doi":"10.1016/j.chemolab.2024.105271","DOIUrl":"10.1016/j.chemolab.2024.105271","url":null,"abstract":"<div><div>The objective of this study is to develop a reliable predictive model for antitumour activity and to identify potential antitumour components in <em>Ganoderma lucidum</em>. Four machine learning models, including Random Forest, were employed to train predictive models for antitumour activity, utilising Morgan fingerprints as molecular descriptors. The most effective model was then employed to predict the chemical composition of <em>Ganoderma lucidum</em>, identifying the four most probable compounds for molecular docking with known TNF-α-related targets. The findings of the study indicate that a Support Vector Machine (SVM) model exhibits an accuracy, F1 score, AUC, and sensitivity of 0.7638, 0.7638, 0.8332, and 0.7621, respectively. The model demonstrated an 80 % accuracy rate in predicting the antitumour activity of 10 FDA-approved drugs. Besides, the model identified 11 components in <em>Ganoderma lucidum</em>, including 3-nitroanisole, with a probability of antitumour activity exceeding 0.5, indicating their potential as antitumour agents. The results of the molecular docking procedure indicated that the four most promising antitumour compounds derived from <em>Ganoderma lucidum</em> exhibited a favourable binding affinity with the TNF-α target. In conclusion, this study incorporated a machine learning prediction step prior to molecular docking, thereby enhancing the reliability of the latter. Furthermore, it identified previously unreported compounds in <em>Ganoderma lucidum</em> with potential antitumour activity, such as 3-nitroanisole.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"255 ","pages":"Article 105271"},"PeriodicalIF":3.7,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652927","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":"Spectra data calibration based on deep residual modeling of independent component regression","authors":"Junhua Zheng ,&nbsp;Zeyu Yang ,&nbsp;Zhiqiang Ge","doi":"10.1016/j.chemolab.2024.105270","DOIUrl":"10.1016/j.chemolab.2024.105270","url":null,"abstract":"<div><div>Independent component regression (ICR) has recently become quite popular in spectra data calibration, due to its advantages in non-Gaussian data modeling and high-order statistics feature extraction. Inspired by the idea of deep learning, this paper extends the basic ICR model to the deep form by introducing a layer-wise residual learning strategy. Based on the residual information generated from last layer of the deep learning model, more and more different patterns of independent components can be extracted layer-by-layer. Then, a further information compression step is taken to combine and also to condense those independent components obtained from different layers of the deep model. Two detailed benchmark case studies are implemented to evaluate the calibration performance of the develop model, based on which the effectiveness of both layer-by-layer component extraction and further information compression are well confirmed.</div></div>","PeriodicalId":9774,"journal":{"name":"Chemometrics and Intelligent Laboratory Systems","volume":"255 ","pages":"Article 105270"},"PeriodicalIF":3.7,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652891","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}