Development of a novel green tea quality roadmap and the complex sensory-associated characteristics exploration using rapid near-infrared spectroscopy technology

Several Feature Extraction Methods Combined with Near-Infrared Spectroscopy for Identifying the Geographical Origins of Milk

Milk is a kind of dairy product with high nutritive value. Tracing the origin of milk can uphold the interests of consumers as well as the stability of the dairy market. In this study, a fuzzy direct linear discriminant analysis (FDLDA) is proposed to extract the near-infrared spectral information of milk by combining fuzzy set theory with direct linear discriminant analysis (DLDA). First, spectral data of the milk samples were collected by a portable NIR spectrometer. Then, the data were preprocessed by Savitzky-Golay (SG) and standard normal variables (SNV) to reduce noise, and the dimensionality of the spectral data was decreased by principal component analysis (PCA). Furthermore, linear discriminant analysis (LDA), DLDA, and FDLDA were employed to transform the spectral data into feature space. Finally, the k-nearest neighbor (KNN) classifier, extreme learning machine (ELM) and naïve Bayes classifier were used for classification. The results of the study showed that the classification accuracy of FDLDA was higher than DLDA when the KNN classifier was used. The highest recognition accuracy of FDLDA, DLDA, and LDA could reach 97.33%, 94.67%, and 94.67%. The classification accuracy of FDLDA was also higher than DLDA when using ELM and naïve Bayes classifiers, but the highest recognition accuracy was 88.24% and 92.00%, respectively. Therefore, the KNN classifier outperformed the ELM and naïve Bayes classifiers. This study demonstrated that combining FDLDA, DLDA, and LDA with NIR spectroscopy as an effective method for determining the origin of milk.

Co-expression of metabolites and sensory attributes through weighted correlation network analysis to explore flavor-contributing factors in various Pyrus spp. Cultivars

Flavor profiles of various Pyrus spp. cultivars exhibit significant variations, yet the underlying flavor-contributing factors remain elusive. In this investigation, a comprehensive approach encompassing metabolomics analysis, volatile fingerprint analysis, and descriptive sensory analysis was employed to elucidate the flavor disparities among Nanguoli, Korla fragrant pear, and Qiuyueli cultivars and uncover potential flavor contributor. The study comprehensively characterized the categories and concentrations of nonvolatile and volatile metabolites, and 925 metabolites were identified. Flavonoids and esters dominated the highest cumulative response, respectively. Utilizing weighted correlation network analysis (WGCNA), seven highly correlated modules were identified, yielding 407 pivotal metabolites. Further correlation analysis of the differential substances provided potential flavor constituents strongly associated with various sensory attributes; taste factors had a certain association with olfactory characteristics. Our findings demonstrated the manifestation of flavor was a result of the synergistic effect of various compounds; evaluation olfactory flavor necessitated a comprehensive consideration of taste substances.

Estimating the sensory-associated metabolites profiling of matcha based on PDO attributes as elucidated by NIRS and MS approaches

Matcha has been globally valued by consumers for its distinctive fragrance and flavor since ancient times. Currently, the protected designation of origin (PDO) certified matcha, characterized by unique sensory attributes, has garnered renewed interest from consumers and the industry. Given the challenges associated with assessing sensory perceptions, the origin of PDO-certified matcha samples from Guizhou was determined using NIRS and LC-MS platforms. Notably, the accuracy of our established attribute models, based on informative wavelengths selected by the CARS-PLS method, exceeds 0.9 for five sensory attributes, particularly the particle homogeneity attribute (with a validation correlation coefficient of 0.9668). Moreover, an LC-MS method was utilized to analyze non-target matcha metabolites to identify the primary flavor compounds associated with each flavor attribute and to pinpoint the key constituents responsible for variations in grade and flavor intensity. Additionally, high three-way intercorrelations between descriptive sensory attributes, metabolites, and the selected informative wavelengths were observed through network analysis, with correlation coefficients calculated to quantify these relationships. In this research, the integration of matcha chemical composition and sensory panel data was utilized to develop predictive models for assessing the flavor profile of matcha based on its chemical properties.

Mid-infrared spectroscopy for the rapid quantification of eucalyptus oil adulteration in Australian tea tree oil (Melaleuca alternifolia)

Essential oil distilled from Melaleuca alternifolialeaves, commonly known as tea tree oil, is well known for its biological activity, principally its antimicrobial properties. However, many samples are adulterated with other, cheaper essential oils such as eucalyptus oil. Current methods of detecting such adulteration are costly and time-consuming, making them unsuitable for rapid authentication screening. This study investigated the use of mid-infrared (MIR) spectroscopy for detecting and quantifying the level of eucalyptus oil adulteration in spiked samples of pure Australian tea tree oil. To confirm the authenticity of the tea tree oil samples, GC-MS analysis was used to profile 37 of the main volatile constituents present, demonstrating that the samples conformed to ISO specifications. Three chemometric regression techniques (PLSR, PCR and SVR) were trialled on the MIR spectra, along with a variety of pre-processing techniques. The best-performing full-wavelength PLSR model showed excellent prediction of eucalyptus oil content, with an R²_CV of 0.999 and RMSECV of 1.08 % v/v. The RMSECV could be further improved to 0.82 % v/v through a moving window wavenumber optimisation process. The results suggest that MIR spectroscopy combined with PLSR can be used to predict eucalyptus oil adulteration in Australian tea tree oil samples with a high level of accuracy.