Search

This study proposes an efficient method for monitoring the submerged fermentation process of Tremella fuciformis (T. fuciformis) by integrating electronic nose (e-nose), electronic tongue (e-tongue), and colorimeter sensors using a data fusion strategy. Chemometrics was employed to establish qualitative identification and quantitative prediction models. The Pearson correlation analysis was applied to extract features from the e-nose and tongue sensor arrays. The optimal sensor arrays for monitoring the submerged fermentation process of T. fuciformis were obtained, and four different data fusion methods were developed by incorporating the colorimeter data features. To achieve qualitative identification, the physicochemical data and principal component analysis (PCA) results were utilized to determine three stages of the fermentation process. The fusion signal based on full features proved to be the optimal data fusion method, exhibiting the highest accuracy across different models. Notably, random forest (RF) was shown to be the most accurate pattern recognition method in this paper. For quantitative prediction, partial least squares regression (PLSR) and support vector regression (SVR) were employed to predict the sugar content and dry cell weight during fermentation. The best respective predictive R2 values for reducing sugar, tremella polysaccharide and dry cell weight were found to be 0.965, 0.988, and 0.970. Furthermore, due to its ability to capture nonlinear data relationships, SVR had superior performance in prediction modeling than PLSR. The results demonstrated that the combination of electronic sensor fusion signals and chemometrics provided a promising method for effectively monitoring T. fuciformis fermentation. [Display omitted] • A data fusion system based on optimal sensor array was developed. • The fermentation process was identified by physicochemical parameter results. • The visualization results of PCA confirmed the fermentation process. • Multiple feature sets were evaluated by DT、RF、SVC for effective classification. • SVR and PLSR successfully predicted crucial physicochemical parameters. [ABSTRACT FROM AUTHOR]

The aim of this work was to investigate the applicability of e-senses (electronic nose, electronic tongue and electronic eye) for the characterization of edible olive oils (extra virgin, olive and pomace) and for the assessment of extra virgin olive oil and olive oil quality decay during storage at different temperatures. In order to obtain a complete description of oil samples, physico-chemical analyses on quality and nutritional parameters were also performed. Data were processed by PCA and a targeted data processing flow-sheet has been applied to physico-chemical and e-senses dataset starting from data pre-processing introducing an innovative normalization method, called t0 centering. On e-senses data a powerful mid-level data fusion approach has been employed to extract relevant information from different analytical sources combining their individual contributions. On physico-chemical data, an alternative approach for grouping extra virgin olive oil and olive oil samples on the basis of their freshness was applied and two classes were identified: fresh and oxidized. A k-NN classification rule was developed to test the performance of e-senses to classify samples in the two classes of freshness and the average value of correctly classified samples was 94%. Results demonstrated that the combined application of e-senses and the innovative data processing strategy allows to characterize edible olive oils of different categories on the basis of their sensorial properties and also to follow the evolution during storage of extra-virgin olive oil and olive oil sensorial properties thus assessing the quality decay of oils. [ABSTRACT FROM AUTHOR]

Electronic tongues (ETs) are multisensor systems based on the coupling of different sensors with high stability and cross-sensitivity towards different species, along with an appropriate chemometric tool that allow the classification of samples and/or the quantification of selected analytes. Despite the great advantages that the data processing stage may offer (e.g. allowing to offset matrix effects from the sample itself, or to resolve the presence of interferences, drifts or non-linearities obtained with the sensors), the stability of the sensors response together with its varied cross-sensitivity are of utmost importance. In this regard, despite the progress that has been made over the last years and the increasing number of publications dealing with ETs, researchers have focused mainly on the developed application itself or on the data treatment stage, but very few report on the choice of the sensors. In this direction, herein we explore on the development of a simple methodology that allows the a priori selection of the optimal sensor array to carry out a specific application. Concretely, the usage of principal component analysis (PCA) in combination with different clustering indices is proposed as a simple approach to simultaneously assess the cross-response and the reproducibility of the different sensors in a single step, what allows to define which sensors may constitute an ET array for a given qualitative or quantitative application. The above ideas have been applied to the study case of aromatic essential oils. [Display omitted] • A simple methodology for the a priori tailored selection of optimal sensor arrays is proposed. • Selection is based on the combination of PCA with different clustering indexes. • Clustering improvement for qualitative applications is visually and numerically confirmed. • PLS was used to assess also the improvement in quantitative applications. • The same approach can be extrapolated to electronic noses or other array-based systems. [ABSTRACT FROM AUTHOR]

The diagnosis of cancer and other diseases using data from non-specific sensors – such as the electronic tongues (e-tongues) - is challenging owing to the lack of selectivity, in addition to the variability of biological samples. In this study, we demonstrate that impedance data obtained with an e-tongue in saliva samples can be used to diagnose cancer in the mouth. Data taken with a single-response microfluidic e-tongue applied to the saliva of 27 individuals were treated with multidimensional projection techniques and non-supervised and supervised machine learning algorithms. The distinction between healthy individuals and patients with cancer on the floor of mouth or oral cavity could only be made with supervised learning. Accuracy above 80% was obtained for the binary classification (YES or NO for cancer) using a Support Vector Machine (SVM) with radial basis function kernel and Random Forest. In the classification considering the type of cancer, the accuracy dropped to ca. 70%. The accuracy tended to increase when clinical information such as alcohol consumption was used in conjunction with the e-tongue data. With the random forest algorithm, the rules to explain the diagnosis could be identified using the concept of Multidimensional Calibration Space. Since the training of the machine learning algorithms is believed to be more efficient when the data of a larger number of patients are employed, the approach presented here is promising for computer-assisted diagnosis. [Display omitted] • First use of an impedimetric electronic tongue (e-tongue) for the diagnosis of cancer. • E-tongue data were used in cancer diagnosis without needing a specific biomarker. • From saliva samples our e-tongue translated information of clinical importance. • Multidimensional calibration space created rules to explain the diagnosis results. [ABSTRACT FROM AUTHOR]