Partial Least Squares Discriminant Analysis Research Articles

Improving skin cancer detection by Raman spectroscopy using convolutional neural networks and data augmentation.

Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study is to determine if skin cancer detection can be further improved by combining deep neural networks and Raman spectroscopy. Raman spectra of 731 skin lesions were included in this study, containing 340 cancerous and precancerous lesions (melanoma, basal cell carcinoma, squamous cell carcinoma and actinic keratosis) and 391 benign lesions (melanocytic nevus and seborrheic keratosis). One-dimensional convolutional neural networks (1D-CNN) were developed for Raman spectral classification. The stratified samples were divided randomly into training (70%), validation (10%) and test set (20%), and were repeated 56 times using parallel computing. Different data augmentation strategies were implemented for the training dataset, including added random noise, spectral shift, spectral combination and artificially synthesized Raman spectra using one-dimensional generative adversarial networks (1D-GAN). The area under the receiver operating characteristic curve (ROC AUC) was used as a measure of the diagnostic performance. Conventional machine learning approaches, including partial least squares for discriminant analysis (PLS-DA), principal component and linear discriminant analysis (PC-LDA), support vector machine (SVM), and logistic regression (LR) were evaluated for comparison with the same data splitting scheme as the 1D-CNN. The ROC AUC of the test dataset based on the original training spectra were 0.886±0.022 (1D-CNN), 0.870±0.028 (PLS-DA), 0.875±0.033 (PC-LDA), 0.864±0.027 (SVM), and 0.525±0.045 (LR), which were improved to 0.909±0.021 (1D-CNN), 0.899±0.022 (PLS-DA), 0.895±0.022 (PC-LDA), 0.901±0.020 (SVM), and 0.897±0.021 (LR) respectively after augmentation of the training dataset (p<0.0001, Wilcoxon test). Paired analyses of 1D-CNN with conventional machine learning approaches showed that 1D-CNN had a 1-3% improvement (p<0.001, Wilcoxon test). Data augmentation not only improved the performance of both deep neural networks and conventional machine learning techniques by 2-4%, but also improved the performance of the models on spectra with higher noise or spectral shifting. Convolutional neural networks slightly outperformed conventional machine learning approaches for skin cancer detection by Raman spectroscopy.

Near-infrared spectroscopy for the quality control of Sarassum fusiforme: Prediction of antioxidant capability of Sarassum fusiforme at different growth stages

The healthy benefits of seaweed have increased its market demand in recent times. Quality control is crucial for seaweed to ensure the customers’ interest and the sustainable development of seaweed farming industry. This study developed a quality control method for seaweed Sargassum fusiforme, rapid and simple, using near-infrared spectroscopy (NIR) and chemometrics for the prediction of antioxidant capacity of S. fusiforme from different growth stages, S. fusiforme was distinguished according to growth stage by partial least squares-discriminant analysis (PLS-DA) and particle swarm optimization-support vector machine (PSO-SVM). The antioxidant properties including 2,2′-azinobis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS) scavenging capacity, 2,2-diphenyl-1-picrylhydrazyl (DPPH) scavenging capacity, and ferric reducing antioxidant power (FRAP) were quantified using competitive adaptive reweighted sampling (CARS)-PLS model. Based on the spectra data preprocessed by multiplicative scatter and standard normal variate methods, the PSO-SVM models can accurately identify the growth stage of all S. fusiforme samples. The CARS-PLS models exhibited good performance in predicting the antioxidant capacity of S. fusiforme, with coefficient of determination (RP2) and root mean square error (RMSEP) values in the independent prediction sets reaching 0.9778 and 0.4018 % for ABTS, 0.9414 and 2.0795 % for DPPH, and 0.9763 and 2.4386 μmol L−1 for FRAP, respectively. The quality and market price of S. fusiforme should increase in the order of maturation < growth < seedling regarding the antioxidant property. The overall results indicated that the NIR spectroscopy accompanied by chemometrics can assist for the quality control of S. fusiforme in a more rapid and simple manner. This study also provided a customer-oriented concept of seaweed quality grading based on deep insight into the antioxidant capability of S. fusiforme at different growth stages, which is highly valuable for precise quality control and standardization of seaweed market.

An efficient method for tracing the geographic origin of Enshi Yulu fresh tea leaves based on near infrared spectroscopy combined with synergy interval PLS and genetic algorithm

This paper applies near-infrared spectroscopy (NIRS) and multiple chemometrics to efficiently distinguish the origins of fresh tea leaves. The key components were obtained using the partial least squares discriminant analysis (PLS-DA) method. PLS, synergy interval PLS (siPLS), principal component analysis (PCA), genetic algorithm (GA), and their combination methods were used to establish NIRS non-destructive discrimination models. Then, the practical application was examined using external samples. The study identified nine key components (variable importance for the projection (VIP) > 1): epigallocatechin, epicatechin, total sugar, water extracts, total catechins, gallocatechin gallate, tea polyphenols, gallocatechin, and epigallocatechin gallate. Of the six NIRS models, the siPLS-GA model that used 37 spectral data points produced the best results (Rp2 = 0.9706, RMSEP = 0.0772, RPD = 6.59). This model had a prediction accuracy of 96.67% for the prediction set samples and 93.33% for the external samples. It offers a rapid, precise, and non-invasive approach to monitor and regulate the illicit trade of fresh tea leaves, thereby guaranteeing the authenticity of Enshi Yulu products from the processing source and fostering the long-term prosperity and stability of the Enshi Yulu tea industry.

Identification of 1-Methylnicotinamide as a specific biomarker for the progression of cirrhosis to hepatocellular carcinoma

PurposeHepatocellular carcinoma (HCC) is a prevalent malignant tumor, often arising from hepatitis induced by the hepatitis B virus (HBV) in China. However, effective biomarkers for early diagnosis are lacking, leading to a 5-year overall survival rate of less than 20% among patients with advanced HCC. This study aims to identify serum biomarkers for early HCC diagnosis to enhance patient survival rates.MethodsWe established an independent cohort comprising 27 healthy individuals, 13 patients with HBV-induced cirrhosis, 13 patients with hepatitis B-type HCC, and 8 patients who progressed from cirrhosis to hepatocellular carcinoma during follow-up. Serum metabolic abnormalities during the progression from cirrhosis to HCC were studied using untargeted metabolomics. Liquid chromatography-mass spectrometry-based metabolomics methods characterized the subjects’ serum metabolic profiles. Partial least squares discriminant analysis (PLS-DA) was employed to elucidate metabolic profile changes during the progression from cirrhosis to HCC. Differentially expressed metabolites (DEMs) between cirrhosis and HCC groups were identified using the LIMMA package in the R language. Two machine learning algorithms, Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest Classifier (RF), were used to identify key metabolic biomarkers involved in the progression from cirrhosis to HCC. Key metabolic biomarkers were further validated using targeted metabolomics in a new independent validation cohort comprising 25 healthy individuals and 25 patients with early-stage hepatocellular carcinoma.ResultsA total of 155 serum metabolites were identified, of which 21/54 metabolites exhibited significant changes in HCC patients compared with cirrhosis patients and healthy individuals, respectively. PLS-DA clustering results demonstrated a significant change trend in the serum metabolic profile of patients with HBV-induced cirrhosis during the progression to HCC. Utilizing LASSO regression and RF algorithms, we confirmed 10 key metabolic biomarkers. Notably, 1-Methylnicotinamide (1-MNAM) exhibited a persistent and significant decrease in healthy individuals, cirrhosis, and HCC patients. Moreover, 1-MNAM levels in developing patients were significantly higher during the cirrhosis stage than in the HCC stage. Targeted metabolomic validation in an external cohort further confirmed the good diagnostic performance of 1-MNAM in early HCC detection.ConclusionOur findings imply that 1-MNAM may be a specific biomarker for the progression of cirrhosis to HCC.

The difference analysis of physicochemical indexes and volatile flavor compounds of chili oil prepared from different varieties of chili pepper

Because of its peculiar flavor, chili oil is widely used in all kinds of food and is welcomed by people. Chili pepper is an important raw material affecting its quality, and commercial chili oil needs to meet various production needs, so it needs to be made with different chili peppers. However, the current compounding method mainly relies on the experience of professionals and lacks the basis of objective numerical analysis. In this study, the chroma and capsaicinoids of different chili oils were analyzed, and then the volatile components were determined by gas chromatography-mass spectrometry (GC–MS) and gas chromatography-ion migration spectrometer (GC-IMS) and electronic nose (E-nose). The results showed that Zidantou chili oil had the highest L*, b*, and color intensity (ΔE) (52.76 ± 0.52, 88.72 ± 0.89, and 118.84 ± 1.14), but the color was tended to be greenyellow. Xinyidai chili oil had the highest a* (65.04 ± 0.2). But its b* and L* were relatively low (76.17 ± 0.29 and 45.41 ± 0.16), and the oil was dark red. For capsaicinoids, Xiaomila chili oil had the highest content of capsaicinoids was 2.68 ± 0.07 g/kg, Tianjiao chili oil had the lowest content of capsaicinoids was 0.0044 ± 0.0044 g/kg. Besides, 96 and 54 volatile flavor substances were identified by GC–MS and GC-IMS respectively. And the main volatile flavor substances of chili oil were aldehydes, alcohols, ketones, and esters. A total of 11 key flavor compounds were screened by the relative odor activity value (ROAV). Moguijiao chili oil and Zidantou chili oil had a prominent grass aroma because of hexanal, while Shizhuhong chili oil, Denglongjiao chili oil, Erjingtiao chili oil, and Zhoujiao chili oil had a prominent floral aroma because of 2, 3-butanediol. Chili oils could be well divided into 3 groups by the partial least squares discriminant analysis (PLS-DA). According to the above results, the 10 kinds of chili oil had their own characteristics in color, capsaicinoids and flavor. Based on quantitative physicochemical indicators and flavor substances, the theoretical basis for the compounding of chili oil could be provided to meet the production demand more scientifically and accurately.

Characterization of Near Infrared-Dye Colored Fabrics Using Hyperspectral Imaging.

Near-infrared (NIR) dyes have a unique ability to interact favorably with light in the NIR region, which is particularly interesting where stealth and camouflage are paramount, such as in military uniforms. Characterization of cotton fabric dyed with NIR-absorbing dyes using visible-NIR (Vis-NIR) and short-wave infrared (SWIR) hyperspectral imaging was done. The aim of the study was to discern spectral changes caused by variations in dye concentration and dyeing temperature as these parameters directly influence color- and crocking-fastness of fabrics impacting the camouflage effect. The fabric was dyed at three concentrations (2.5, 5, and 10%) and two dyeing temperatures (55 °C and 85 °C) and principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were performed on the spectra to discriminate the fabrics based on dye concentrations. The PCA models successfully segregated the fabrics based on the dye concentration and dyeing temperature, while PLS-DA models demonstrated classification accuracies between 75 and 100% in the Vis-NIR range. Spectra in the SWIR region could not be used to detect the differences in the concentrations of the NIR dyes. This finding is promising, as it aligns with the objective of creating NIR-dyed camouflage fabrics that remain indistinguishable under varying dye concentrations. These results open possibilities for further exploration in enhancing the stealth capabilities of textiles in military applications.

Matrix-Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrometry Combined with Chemometrics for Protein Profiling and Classification of Boiled and Extruded Quinoa from Conventional and Organic Crops

Quinoa is an Andean crop that stands out as a high-quality protein-rich and gluten-free food. However, its increasing popularity exposes quinoa products to the potential risk of adulteration with cheaper cereals. Consequently, there is a need for novel methodologies to accurately characterize the composition of quinoa, which is influenced not only by the variety type but also by the farming and processing conditions. In this study, we present a rapid and straightforward method based on matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) to generate global fingerprints of quinoa proteins from white quinoa varieties, which were cultivated under conventional and organic farming and processed through boiling and extrusion. The mass spectra of the different protein extracts were processed using the MALDIquant software (version 1.19.3), detecting 49 proteins (with 31 tentatively identified). Intensity values from these proteins were then considered protein fingerprints for multivariate data analysis. Our results revealed reliable partial least squares-discriminant analysis (PLS-DA) classification models for distinguishing between farming and processing conditions, and the detected proteins that were critical for differentiation. They confirm the effectiveness of tracing the agricultural origins and technological treatments of quinoa grains through protein fingerprinting by MALDI-TOF-MS and chemometrics. This untargeted approach offers promising applications in food control and the food-processing industry.

Revealing the effects and mechanism of wine processing on Corni Fructus using chemical characterization integrated with multi-dimensional analyses

Corni fructus (CF) is always subjected to wine processing before prescription in clinic, for an enhancing effect of nourishing liver and kidney. While, the underlying mechanism for this processing on CF remains obscure. In this study, a sensitive ultra-high-performance liquid chromatography mass spectrometry (UPLC-MS/MS) method combined multi-dimensional analyses was established to monitor chemical characterizations of raw and wine-processed CF (WCF) and hence reveal the effects and underlying mechanism of wine processing on CF. As indicated, a total of 216 compounds were tentatively identified, including 98 structurally complex and variable home/hetero-polymers, that were composed of iridoid glucosides, gallic acids, caffeic acid and/or 5-HMF. Interestingly, 53 of these compounds probably characterized potential novel, including 35 iridoid glucosides or their dimers, 9 iridoid glucoside-gallic acid dimers, 7 gallic acids derivatives and 2 gallic acid-caffeic acid dimers, which provides ideas for natural product researchers. Meanwhile, the multi-dimensional analyses including principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA) and linear regression analysis were used to explore the differences between CF and WCF. The results showed that 23 compounds as chemical markers greatly contributing to the distinction were screened out, and 3 of which (7α/β-O-ethyl-morroniside, gallic acid and 5-HMF) in WCF indicated an increasing trend in intensities in relative to those in CF. Additionally, linear regression analysis showed that in WCF 53 compounds exhibited an increasing in intensities, while 132 ones did a decreasing trend, compared with those in CF. As our investigation demonstrated, acetal reaction of morroniside, ester hydrolysis in different organic acid derivatives as well as glycoside bond cleavage during wine processing probably resulted in the distinctions. The findings of this study provide a further understanding of the effect and mechanism of wine processing on CF.

Detection of dried jujube from fresh jujube with different variety and maturity after hot air drying based on hyperspectral imaging technology

The consistent quality of dried jujube has a significant impact on its economic value and consumer acceptance. Currently, the evaluation of dried jujube quality heavily relies on agricultural experts and advanced instrument analysis, which is time costly and is limited by many uncertainties. This study aims to develop a rapid and accurate method for classifying dried jujube produced from fresh jujube at different stages of maturity using hyperspectral imaging. Dried jujube samples were collected from two varieties (Zhongyang Muzao jujube and Yuanling jujube) in Hebei Province, representing two maturity stages: Kimri and Rutab. Linear partial least squares-discriminant analysis (PLS-DA), K-Nearest Neighbor (KNN) and support vector machine (SVM)algorithms were combined with six pre-processing methods. the pretreatment spectra of Area Naturalization (AN), Baseline and Multiple scattering correction (MSC) were selected for further analysis. Successive projections algorithm (SPA), Competitive adaptive reweighted sampling （CARS）and Uninformative Variables Elimination (UVE)were combined to select characteristic variables. A total of 16 characteristic variables were selected by AN-UVE-SPA-SVM model. This model achieved the highest accuracies of 96.0 % and 93.1 % in the final discriminant accuracy for the calibration and Validation sets, respectively. The hardness, soluble solid content (SSC), and moisture of two categories of dried jujube derived from fresh jujube at various stages of maturity were analyzed. The results showed that dried jujube produced from fresh jujubes at the Rutab stage were more likely to exhibit characteristics of high SSC and low hardness' compared to those produced at the Kimri stage. The partial least squares regression (PLSR) model established based on characteristic wavelengths extracted by AN-UVE-SPA showed the best results in predicting SSC and moisture. This study showcased the feasibility of identifying and classifying dried jujube using hyperspectral imaging technology, laying a solid foundation for further research in the rapid non-destructive testing of dried fruits and vegetables through the application of hyperspectral technology.

Sensory and chemical characterization of chestnuts processed in different methods using instrumental analyses and the Check-all-that-apply method

Chestnuts are known for their unique sensory characteristics and nutritional value. However, the changes in their sensory attributes after processing remain unclear. This study used instrumental analysis and sensory evaluation combined with multivariate statistical analysis to investigate the effect of packaging and thermal sterilization procedures on the sensory characteristics of chestnuts. The results showed that the significant variations (p < 0.05) between the different processing methods for chestnuts were revealed via the texture analysis, headspace solid-phase microextraction gas chromatography × gas chromatography-time of flight-mass spectrometry (HS-SPME-GC×GC-TOF-MS), and check-all-that-apply (CATA) sensory evaluation. The packaging had a more significant influence on the sensory quality of the chestnuts than thermal sterilization procedures. The HS-SPME-GC×GC-TOF-MS identified 116 volatile compounds, and the partial least squares-discriminant analysis (PLS-DA) identified the 28 volatile compounds responsible for the similarities and differences among different processing methods of chestnuts. Principal component analysis (PCA) combined with texture analysis and the fingerprint of HS-SPME-GC×GC-TOF-MS, as well as the correspondence analysis (CA) analysis of CATA, could discriminate samples from different packaging of chestnuts. This study provides chestnuts processing recommendations for satisfying consumer preferences and predicting possible related quality changes.

Indirect evaluation of watermelon volatile profile: Detection of subtle changes with e-nose systems

The effectiveness of e-nose systems as high-throughput tools for volatile profiling in watermelon was investigated focusing on discerning subtle changes induced by the use of different rootstocks. Partial Least Square Discriminant Analysis (PLS-DA) models, both GC-MS and e-nose data, demonstrated moderate performance in classification due to nuanced differences among groups (the same F1 hybrid was used as scion). However, PLS-DA biplots revealed a clear correlation between GC-MS and e-nose data. This methodology enabled the e-nose system to identify the effects of specific root-scion combinations compared to non-grafted controls and detect combinations with more variable volatile profiles. Remarkably, the e-nose system identified samples with anomalous volatile profiles, mirroring the capabilities of GC-MS data. Additionally, PLS models were developed to provide reasonably accurate predictions of key compound contents like geranylacetone, (Z)-6-nonen-1-ol, or (Z)-6-nonenal, crucial for watermelon flavor and taste perception. Overall, this study highlights the potential of e-nose systems in discerning nuanced variations in watermelon volatile profiles affecting aroma. Incorporating volatile profile evaluation capabilities using such systems will significantly optimize quality control processes and plant breeding programs.

Continued development of a tool to measure intramuscular fat of Australian lamb carcases in commercial situations

Despite the importance of intramuscular fat (IMF) to eating quality, as yet no methodology has been widely adopted by the whole of industry in Australia to measure it routinely. Thus, a study was conducted to investigate the potential for a Near Infra-Red (NIR) device to predict the IMF content of the loin from spectra collected on the topside which is externally located on a hanging carcase and therefore easily accessible. To this end, NIR spectra were collected from topsides (m. semimembranosus) of 258 lamb carcases over 5 data collections and a sample of muscle was collected from the loin and the topside for IMF determination using a wet chemistry method. Subsequent Partial Least Square (PLS) models suggested the ability to predict the absolute IMF content of loins was poor (R2 = 0.28, RMSE = 1.26), yet there was a moderate ability to predict the IMF content of the topside (R2 = 0.56, RMSE = 0.82). Partial Least Square Discrimination Analysis (PLS-DA) models to classify cuts based on the IMF eating quality threshold of 4.5% yielded better predictive outcomes with accuracies of 66.7% and 76.7% for loin and topside respectively. However, further research to assess the relationship between the IMF of the loin and topside and reduce the impact of differences in overall absorbance between data collections will improve predictive outcomes.

Quality Evaluation of Tripterygium Glycoside Tablets Based on Quantitative Band-Selective 2D 1H-13C HSQC and 1H NMR Fingerprinting.

Tripterygium glycoside tablets (TGTs) are preparations extracted and purified from Tripterygium wilfordii Hook. F and are extensively utilized in the treatment of autoimmune diseases, such as rheumatoid arthritis (RA). However, variations in production processes among manufacturers can lead to challenges in quality control and clinical utilization of TGTs. A band-selective 2D 1H-13C HSQC quantification method was applied for the determination of 13 active ingredients in TGTs. This method was validated following the guidelines of USP-NF 2022. The results demonstrated that the quantitative method exhibited excellent signal resolution, as well as sufficient accuracy, sensitivity, and stability. In addition, the 1H NMR spectra of TGTs from three manufacturers underwent analysis using principal component analysis and orthogonal partial least-squares discriminant analysis. The results revealed significant differences among the TGTs from the three manufacturers, with manufacturer 2 and manufacturer 3 demonstrating superior product consistency compared to manufacturer 1. A quality evaluation system for TGTs was developed based on band-selective 2D 1H-13C HSQC and 1H NMR, encompassing both quality markers and fingerprinting. This system offers reliable approaches and insights for enhancing the quality control of natural products.

Quantification of AICAR and study of metabolic markers after administration.

Objectives: AICAR (5-amino-4-imidazolecarboxyamide ribonucleoside) was reported as the first pharmacological AMPK (adenosine 5'-monophosphate (AMP)-activated protein kinase) activator, and it has been confirmed to exhibit a significant endurance enhancement effect and prohibited for doping by the World Anti-Doping Agency. Due to the fact that the human body can produce such substances, in order to ensure fairness in sports competition, methods for rapid detection and multi-type identification of AICAR drugs taken orally should be established. Methods: to assess AICAR levels, a new rapid, sensitive, efficient, and selective method was reported for the quantitative detection of AICAR in urine using LC-MS/MS. The method was validated for quantitative purposes based on the elemental selectivity, intra- (1.0-15.6%) and inter-day precision (1.3-16.3%), accuracy (99.9-112.8%), matrix effects (88.9-103.6%), recovery (87.4-106.5%), and stability at four different concentrations. The calibration curve was linear over a wide concentration range of 10-10,000 ng mL-1 with a high coefficient of determination (R 2 > 0.998). The limit of detection (LOD) and limit of quantification (LOQ) for the experiment were determined to be 1 and 10 ng mL-1, respectively. Simultaneously, metabolomics analysis was used to obtain the metabolic fingerprint of different populations and biomarkers to distinguish administration cases through partial least squares discriminant analysis (PLS-DA) and a receiver operating characteristic (ROC) curve. Results: the method enables easy quantitation for LC-MS/MS analysis with the best recovery yield maintained, and the method was applied to 122 Asian biological samples with an average concentration of 1310.5 ± 1031.4 ng mL-1. Through drug metabolism research, 734 and 294 variables were extracted for data analysis respectively in the positive and negative ion modes, and more than 100 metabolites with significant up- and down-regulation were found after the test. Conclusions: this research developed a fast, precise, effective, and specific approach for the qualitative and quantitative identification of AICAR in urine. Meanwhile, administration metabolism studies found that there were significant changes in AICAR levels and other compounds, such as PC types PC(18:1/16:0), PC(16:0/18:0), and PC(16:0/16:0), PE types PE(18:0/20:4), and LPE-type 18:1, which could better distinguish samples before and after AICAR administration. The analysis provides a multi-perspective reference for WADA to determine a positive criterion.

Discrepancy on the flavor compound affect the quality of Taiping Houkui tea from different production regions

Taiping Houkui (TPHK) is prevalent green tea in China, its flavor quality is significantly influenced by different production regions. However, the key flavor compounds responsible for these discrepancies remain unclearly. Here, TPHK samples were produced from fresh leaves of ‘Shidacha 2’ cultivar planted in 14 distinct production regions. In 14 TPHK samples, a total of 33 non-volatile compounds were identified and quantified. Partial least-squares discriminant analysis (PLS-DA) reveal that theanine and glutamate were the main umami compounds, caffeine imparted with bitterness, which collectively contributed to the variation in the taste flavor of TPHK across different production regions. Furthermore, the profiles of 51 volatile compounds were determined, integrated PLS-DA with odor activity values of volatiles indicated that linalool (165.7–888.5) and geraniol (11.9–141.4) affecting the floral aroma of TPHK among different production regions. Our findings revealed the critical compounds that contributed to the effect of production regions on flavor quality of TPHK.

Fraud detection in the fishing sector using hyperspectral imaging

Currently, more and more consumers are interested in the quality, safety, and authenticity of food products. The fishing sector is the second food category with the highest risk of fraud and the greatest presence of authentication problems. There are non-destructive, fast and accurate techniques for real-time authentication, with hyperspectral imaging (HSI) standing out among these. In this context, the main aim of this study is to explore the viability of HSI in the visible and near infrared (VIS-NIR) and near infrared (NIR) ranges for the detection of fraud by origin and by non-declaration of the previous freezing process, in anchovies. The spectral pretreatment methods used were the standard normal variate method, the Savitzky-Golay 1st derivate and the Savitzky-Golay 2nd derivate, always followed by mean centering (MC). In addition, the impact of using a previous step of smoothing prior to pretreatment was also evaluated. Two classification algorithms: soft independent modeling of class analogy, and partial least squares discriminant analysis (PLS-DA) were used to build the classification model. After analysis, it was found that the modelling results using the VIS-NIR region were always better than those using the NIR region, and the best performing model was by PLS-DA with a recall of 0.97 for fresh and 0.98 for frozen-thawed anchovies and 0.98 for Cantabrian anchovies and 0.96 for Mediterranean anchovies. One advantage of the model obtained is the ability to classify the anchovies measuring on the skin side of fish without the need for sample preparation. Overall, the results showed that HSI combined with PLS-DA is a favorable solution for rapid, and non-destructive recognition of adulteration regarding freshness and origin in anchovies.

Combined Relaxation Spectra for the Prediction of Meat Quality: A Case Study on Broiler Breast Fillets with the Wooden Breast Condition.

This study evaluated the potential of using combined relaxation (CRelax) spectra within time-domain nuclear magnetic resonance (TD-NMR) measurements to predict meat quality. Broiler fillets affected by different severities of the wooden breast (WB) conditions were used as case-study samples because of the broader ranges of meat-quality variations. Partial least squares regression (PLSR) models were established to predict water-holding capacity (WHC) and meat texture, demonstrating superior CRelax capabilities for predicting meat quality. Additionally, a partial least squares discriminant analysis (PLS-DA) model was developed to predict WB severity based on CRelax spectra. The models exhibited high accuracy in distinguishing normal fillets from those affected by the WB condition and demonstrated competitive performance in classifying WB severity. This research contributes innovative insights into advanced spectroscopic techniques for comprehensive meat-quality evaluation, with implications for enhancing precision in meat applications.

AI-based mining of biomedical literature: Applications for drug repurposing for the treatment of dementia.

This manuscript outlines our project involving the application of AGATHA, an AI-based literature mining tool, to discover drugs with the potential for repurposing in the context of neurocognitive disorders. The primary objective is to identify connections between approved medications and specific health conditions through advanced statistical analysis, including techniques like Partial Least Squares Discriminant Analysis (PLSDA) and unsupervised clustering. The methodology involves grouping scientific terms related to different health conditions and genes, followed by building discrimination models to extract lists of disease-specific genes. These genes are then analyzed through pathway analysis to select candidates for drug repurposing.

Advanced chemometric methodologies on single shot hyphenated LIBS data for rapid and reliable characterization of plastic classes

BackgroundPlastic Solid Waste (PSW) sorting is a procedure of paramount importance in the mechanical recycling process of plastics waste. The major limitation of the techniques relying on physical properties of plastics is the time taken for analysis and poor accuracy. Spectroscopy has been shown to be a suitable method in plastic sorting due to its atomic and molecular characterization capabilities, and ability to give results in very short time scales. However, for practical applications it is essential to translate this technique into an automatic technology, by combining it with advanced chemometric tools which can give observer independent judgement. ResultsThe indigenously developed bi-model Laser Induced Breakdown Spectroscopy (LIBS)-Raman system with single source and single detector can record the LIBS-Raman spectral signals in single-shot mode in a total time frame of 20 ms. Out of the combinations of Principal Component Analysis (PCA) and Partial Least Squares (PLS) with Logistic Regression (LR), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), and Partial Least Squares-Discriminant Analysis (PLS-DA) based classifiers, the PLS-DA based model showed the maximum classification accuracy with 95 % based on LIBS data and 100 % based on Raman data. The reliability of the model was assessed using 4-fold cross-validation which showed a sensitivity of 90.28 % and specificity of 98.29 % for predictions based on LIBS data, and 99 % sensitivity and 99.82 % specificity for predictions relying on Raman data. SignificanceThe results show how the combination of multimodal spectroscopy with chemometric analysis enhances the applicability of spectroscopic techniques for plastic sorting. The classification model successfully classified seven types of post-consumer plastic samples based on combined LIBS and Raman data. With the home-built software for automated prediction, the system takes less than a second to predict the plastic type illustrating the potential of the method for translation to regular routine industrial applications.

Quality Evaluation of Calyx of Diospyros Kaki Thunb. by Ultra-High Performance Liquid Chromatography – Mass Spectrometry (UHPLC-MS) Profiling and Fingerprinting and Chemometrics

Calyx of Diospyros kaki Thunb. (DKC), a famous traditional Chinese medicine, has a remarkable therapeutic effect on hiccups but its quality control is insufficient. In the study, a method was established by ultra-high performance liquid chromatography – mass spectrometry (UHPLC-MS). 17 peaks were observed in the mass spectrum and their chemical structures were preliminarily identified, including 3 organic acids and 14 flavonoids. The reference UHPLC fingerprint was built based on 25 batches of DKC samples, containing 15 common peaks. 11 batches of DKC samples from Guangxi and Hebei were superior to other 14 batches, sharing high similarity from 0.966 to 0.988 with the reference fingerprint. The chemometric methods, including hierarchical clustering analysis (HCA), principal component analysis (PCA), and partial least squares discriminant analysis (PLS-DA), were used to distinguish the DKC samples. The recognition methods clearly distinguished the medicinal materials from different regions, indicating that the main chemical constituents leading to the differences were protocatechuic acid, 4-hydroxybenzoic acid, rutin (presumptive result based on MS data), isoquercitrin, kaempferol, and hyperoside. The established fingerprint method is simple, rapid, repeatable, and specific and provides a reference for the quality evaluation of DKC.