E-nose Data Research Articles

Characterization of pasteurized milk spoilage by electronic nose in relation to its selected quality parameters

Pasteurized fresh milk requires an accurate estimation of shelf life under various conditions to minimize the risk of spoilage and product losses. Milk samples were stored for 56 h in an oven at 25°C and for 15 days in a refrigerator at 4°C. Samples were analyzed using an electronic nose (e-nose), total bacterial count, titratable acidity and pH to determine the quality of milk. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used to analyze e-nose data of milk stored at 25°C, and 4°C. A clear shift in quality was identified by the e-nose, which also appeared in the total bacterial count after 24 h and 12 days for storage at 25 and 4°C, respectively. On the other hand, titratable acidity exceeded the normal limits of 0.14 % - 0.21 % after 24 h for storage at 25°C (0.247 ± 0.006 %) and after 15 days for storage at 4°C (0.25 ± 0.01 %). If pH was a good indicator of quality for samples stored at 25°C, it showed no clear trends for samples stored at 4°C. Based on the microbial count data and e-nose output, the milk had a shelf life of 0.3 day (i.e. 8 h) when stored at 25°C. Shelf life was extended to 9 days when stored at 4°C.

Characterization of pasteurized milk spoilage by electronic nose in relation to its selected quality parameters

Pasteurized fresh milk requires an accurate estimation of shelf life under various conditions to minimize the risk of spoilage and product losses. Milk samples were stored for 56 h in an oven at 25°C and for 15 days in a refrigerator at 4°C. Samples were analyzed using an electronic nose (e-nose), total bacterial count, titratable acidity and pH to determine the quality of milk. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used to analyze e-nose data of milk stored at 25°C, and 4°C. A clear shift in quality was identified by the e-nose, which also appeared in the total bacterial count after 24 h and 12 days for storage at 25 and 4°C, respectively. On the other hand, titratable acidity exceeded the normal limits of 0.14 % - 0.21 % after 24 h for storage at 25°C (0.247 ± 0.006 %) and after 15 days for storage at 4°C (0.25 ± 0.01 %). If pH was a good indicator of quality for samples stored at 25°C, it showed no clear trends for samples stored at 4°C. Based on the microbial count data and e-nose output, the milk had a shelf life of 0.3 day (i.e. 8 h) when stored at 25°C. Shelf life was extended to 9 days when stored at 4°C.

Comparison of different processing approaches by SVM and RF on HS-MS eNose and NIR Spectrometry data for the discrimination of gasoline samples

In the quality control of flammable and combustible liquids, such as gasoline, both rapid analysis and automated data processing are of great importance from an economical viewpoint for the petroleum industry. The present work aims to evaluate the chemometric tools to be applied on the Headspace Mass Spectrometry (HS-MS eNose) and Near-Infrared Spectroscopy (NIRS) results to discriminate gasoline according to their Research Octane Number (RON). For this purpose, data from a total of 50 gasoline samples of two types of RON-95 and 98-analyzed by the two above-mentioned techniques were studied. The HS-MS eNose and NIRS data were combined with non-supervised exploratory techniques, such as Hierarchical Cluster Analysis (HCA), as well as other supervised classification techniques, namely Support Vector Machine (SVM) and Random Forest (RF). For supervised classification, the low-level data fusion was additionally applied to evaluate if the combined use of the data increases the scope of relevant information. The HCA results showed a clear clustering trend of the gasoline samples according to their RON with HS-MS eNose data. SVM in combination with 5-Fold Cross-Validation successfully classified 100% of the samples with the HS-MS eNose data set. The RF algorithm in combination with 5-Fold Cross-Validation achieved the best accuracy rate for the test set with the low-level data fusion system. Furthermore, it allowed us to identify the most important features that could define the differences between RON 95 and RON 98 gasoline. On the other hand, using the HS-MS eNose and NIRS low-level data fusion reached better results than those obtained using NIRS data individually, with accuracy rates of 100% in both SVM and RF performances with the test set. In general, the performance of the SVM and RF algorithms was found to be similar.

Electronic-Nose as Non-destructive Tool to Discriminate "Ferrovia" Sweet Cherries Cold Stored in Air or Packed in High CO2 Modified Atmospheres.

This study aimed to explore the applicability of electronic-nose (E-nose) as a rapid method in discriminating samples of sweet cherry cv “Ferrovia” stored in high-CO2 (16% O2 + 20% CO2 + 64% N2) or air (control) up to 21 days. Projection to Latent Structures (PLS) methods applied to E-nose data showed that fresh fruit and the packaged or unpackaged samples can be distinguished, according to both the storage condition and the storage days. Moreover, a correlation analysis between E-nose sensors and 45 volatile compounds were overall, obtained from all the investigated sweet cherry samples by Headspace Solid-Phase Microextraction (HS SPME) coupled to Gas Chromatography-Mass Spectrometry (GC-MS). These methods allowed to associate samples with a specific flavour profile to one or more E-nose sensors. Finally, quality attributes (visual quality, colour, firmness, antioxidant activity, total phenols, and sugar content) were assessed during storage. Among these, visual quality and berry deformation resulted affected by storage conditions, showing that high-CO2 treatment better preserved the fruit quality than control.

A data fusion approach on confocal Raman microspectroscopy and electronic nose for quantitative evaluation of pesticide residue in tea

Pesticide residue in tea is always known as a major concern in the process of tea quality assessment, as it poses potential risks to health even at very low levels. To explore the best approach for quantitative detection of pesticide residue in tea, electronic nose (e-nose) and confocal Raman microspectroscopy (CRM) were applied to prove their capability in the determination of chlorpyrifos concentration. Partial least square (PLS) regression was developed based on the e-nose and spectral variables separately and the fusion of the variables into a single array, to analyze the performance of the variable selection algorithms and use the complementary data acquired by e-nose and CRM sensing technologies. Support vector machine (SVM) and artificial neural network (ANN) models were then developed to the individual datasets and the fused dataset, to correlate the signals obtained by both technologies, with the concentration of pesticide residues measured in samples by reference analytical method. The detection model developed based on the data fusion outperformed those based on e-nose and CRM separately, and the result showed that both technologies had a major role in predicting the contamination of pesticide residue. The best prediction performance was derived with 32 effective variables selected from the fusion dataset by the ANN model, with the optimum value of RMSEP (0.0135) and R 2 p value of 0.973. This research demonstrated the high potential to combine e-nose and CRM data as an alternative approach for determining pesticide residues in tea processing, especially coupled with efficient chemometric strategies. • The concentration of pesticide residue in tea was evaluated non-destructively. • Electronic nose and confocal Raman microspectroscopy were used for the prediction. • Spectra and aroma data had a good capability for fast safety control of pesticides. • Data fusion with selected variables improved the prediction performances.

HS-SPME-GC-MS and Electronic Nose Reveal Differences in the Volatile Profiles of Hedychium Flowers.

Floral fragrance is one of the most important characteristics of ornamental plants and plays a pivotal role in plant lifespan such as pollinator attraction, pest repelling, and protection against abiotic and biotic stresses. However, the precise determination of floral fragrance is limited. In the present study, the floral volatile compounds of six Hedychium accessions exhibiting from faint to highly fragrant were comparatively analyzed via gas chromatography–mass spectrometry (GC–MS) and Electronic nose (E-nose). A total of 42 volatile compounds were identified through GC–MS analysis, including monoterpenoids (18 compounds), sesquiterpenoids (12), benzenoids/phenylpropanoids (8), fatty acid derivatives (2), and others (2). In Hedychium coronarium ‘ZS’, H. forrestii ‘Gaoling’, H. ‘Jin’, H. ‘Caixia’, and H. ‘Zhaoxia’, monoterpenoids were abundant, while sesquiterpenoids were found in large quantities in H. coccineum ‘KMH’. Hierarchical clustering analysis (HCA) divided the 42 volatile compounds into four different groups (I, II, III, IV), and Spearman correlation analysis showed these compounds to have different degrees of correlation. The E-nose was able to group the different accessions in the principal component analysis (PCA) corresponding to scent intensity. Furthermore, the pattern-recognition findings confirmed that the E-nose data validated the GC–MS results. The partial least squares (PLS) analysis between floral volatile compounds and sensors suggested that specific sensors were highly sensitive to terpenoids. In short, the E-nose is proficient in discriminating Hedychium accessions of different volatile profiles in both quantitative and qualitative aspects, offering an accurate and rapid reference technique for future applications.

Early Detection of Aphid Infestation and Insect-Plant Interaction Assessment in Wheat Using a Low-Cost Electronic Nose (E-Nose), Near-Infrared Spectroscopy and Machine Learning Modeling.

Advances in early insect detection have been reported using digital technologies through camera systems, sensor networks, and remote sensing coupled with machine learning (ML) modeling. However, up to date, there is no cost-effective system to monitor insect presence accurately and insect-plant interactions. This paper presents results on the implementation of near-infrared spectroscopy (NIR) and a low-cost electronic nose (e-nose) coupled with machine learning. Several artificial neural network (ANN) models were developed based on classification to detect the level of infestation and regression to predict insect numbers for both e-nose and NIR inputs, and plant physiological response based on e-nose to predict photosynthesis rate (A), transpiration (E) and stomatal conductance (gs). Results showed high accuracy for classification models ranging within 96.5–99.3% for NIR and between 94.2–99.2% using e-nose data as inputs. For regression models, high correlation coefficients were obtained for physiological parameters (gs, E and A) using e-nose data from all samples as inputs (R = 0.86) and R = 0.94 considering only control plants (no insect presence). Finally, R = 0.97 for NIR and R = 0.99 for e-nose data as inputs were obtained to predict number of insects. Performances for all models developed showed no signs of overfitting. In this paper, a field-based system using unmanned aerial vehicles with the e-nose as payload was proposed and described for deployment of ML models to aid growers in pest management practices.

E-Nose Discrimination of Abnormal Fermentations in Spanish-Style Green Olives.

Current legislation in Spain indicates that table olives must be free of off-odors and off-flavors and without symptoms of ongoing alteration or abnormal fermentations. In this regard, the International Olive Council (IOC) has developed a protocol for the sensory classification of table olives according to the intensity of the predominantly perceived defect (PPD). An electronic nose (e-nose) was used to assess the abnormal fermentation defects of Spanish-style table olives that were previously classified by a tasting panel according to the IOC protocol, namely zapateria, butyric, putrid, and musty or humidity. When olives with different defects were mixed, the putrid defect had the greatest sensory impact on the others, while the butyric defect had the least sensory dominance. A total of 49 volatile compounds were identified by gas chromatography, and each defect was characterized by a specific profile. The e-nose data were analyzed using principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA). The different defects were clearly separated from each other and from the control treatment, independently of PPD intensity. Moreover, the e-nose differentiated control olives from table olives with combined sensory defects despite the dilution effect resulting from the combination. These results demonstrate that e-nose can be used as an olfactory sensor for the organoleptic classification of table olives and can successfully support the tasting panel.

Discrimination of COPD and lung cancer from controls through breath analysis using a self-developed e-nose

This work details the application of a metal oxide semiconductor (MOS) sensor based electronic nose (e-nose) system in the discrimination of lung cancer and chronic obstructive pulmonary disease (COPD) from healthy controls. The sensor array integrated with supervised classification algorithms was able to detect and classify exhaled breath samples from healthy controls, patients with COPD, and lung cancer by recognizing the amount of volatile organic compounds present in it. This paper details the e-nose design, participant selection, sampling methods, and data analysis. The clinical feasibility of the system was checked in 32 lung cancer patients, 38 COPD patients, and 72 healthy controls including smokers and non-smokers. One of the advantages of the equipment design was portability and robustness since the system was conditioned with elements that allowed its easy movement. In the discrimination of lung cancer from controls, the k-nearest neighbors gave an acceptable accuracy, sensitivity, and specificity of 91.3%, 84.4%, and 94.4% respectively. The support vector machine gave better results for COPD discrimination from controls with 90.9% accuracy, 81.6% sensitivity, and 95.8% specificity. Even though the attained results were good, further examinations are essential to enhance the sensor array system, investigate the long-run reproducibility, repeatability, and enlarge its relevancy.

Channel Attention Convolutional Neural Network for Chinese Baijiu Detection With E-Nose

Electronic nose (E-nose) plays an important role in the detection of Chinese baijiu, which is an alcoholic beverage of high reputation. However, traditional e-nose data processing relies on manual selection of features with complicated preprocessing steps. The production process of baijiu with the same flavor is similar, but due to the limited number of sensors, it is not easy to quickly extract specific features with e-nose. To overcome this shortcoming, we propose a novel method based on Channel Attention Convolutional Neural Network (CA-CNN) for authenticity identification of Chinese baijiu. The underlying channel attention module analyzes the dependencies between channels and learning weights in order to improve the detection accuracy. In particular, we evaluated the comprehensive performance of the system by comparing it with traditional machine learning methods. The prediction accuracy of the CA-CNN (98.53%) is better than the back-propagation artificial neural network (BP-ANN) (90.83%), the support vector machine (SVM) (84.5%), and the random forest (RF) (86.867%). The experiment results show that the CA-CNN has reasonable reliability, stability and good prediction performance in the quality classification of Chinese baijiu. It provides an effective reference method for the standardization of baijiu quality inspection.

Evaluation of flavor characteristics of bacon smoked with different woodchips by HS-SPME-GC-MS combined with an electronic tongue and electronic nose

This study investigated the effects of different woodchip types (beech, oak, pear, and apple) on the volatile compounds and sensory characteristics of smoked bacon. The volatile compounds were influenced by woodchip types and the total content of ketones and phenols in pear-smoked bacon were higher than in bacon smoked with other woodchips (P < 0.05). The E-tongue combined with E-nose can effectively distinguish the difference in the flavor of bacon smoked with different woodchip types by the signal intensities. Sensory analysis showed that smoking increased bacon's redness, saltiness, and smoky flavor compared with the control (unsmoked bacon) (P < 0.05) and it had little impact on off-odor (P > 0.05). Correlation analysis showed that the E-nose and E-tongue data were highly correlated with contents of alcohols, aldehydes, and ketones. This study revealed that the different smoked materials greatly influenced the flavor and sensory properties of bacon.

Lung cancer detection via breath by electronic nose enhanced with a sparse group feature selection approach

In the diagnosis of lung cancer, electronic nose (E-nose) has attracted much attention by detecting the volatile organic compounds (VOCs) in exhaled breath. In this research, an E-nose platform with a novel thermal desorption preconcentration subsystem is designed to verify whether analyzing VOCs can reliably differentiate lung cancer patients from healthy individuals and patients with benign pulmonary diseases. To this end, total 87 subjects (46 patients with lung cancer, 36 healthy volunteers and 5 patients with benign pulmonary diseases) are enrolled for the sensor array data collection. 13 composite features are extracted from each sensor, and some classical classifiers are established to demonstrate the feasibility of identifying lung cancer patients through VOCs. To improve the performance, considering the inherent characteristics of E-nose data, a sparse group feature selection (FS) method is applied to the raw sensor array data. It is observed that the FS method can reduce data dimensionality and improve the classification performance significantly. Statistical analysis for the effect of age and smoking habit on classification results shows that no significant influences are found.

Drift Compensation on Electronic Nose Data Relevant to the Monitoring of Odorous Emissions from a Landfill by Opls

Nowadays electronic noses (e-noses) are often prescribed in the permits of new or existing plants for a continuous monitoring of ambient air quality at receptors or plant fenceline. To do this, the e-nose has to guarantee a stable classification and quantification performance over time. However, the problem of drift (i.e., progressive deviation over time of sensor responses) becomes critical in the case of continuous odour monitoring in the field. Thus, specific strategies for drift compensation need to be defined. This paper focuses on the development a specific drift correction model based on OPLS to mitigate drift effects on e-nose data relevant to three olfactometric campaigns carried out at a landfill over three years. The paper aims to reduce costs associated to the recalibration of the decisional model to be performed before every seasonal e-nose monitoring campaign prescribed in the permit of the landfill. The OPLS model was built on the data collected in the first two campaigns, while data relevant to the most recent campaign were used to test its efficacy in compensating drift. The results achieved pointed out the potentialities of the OPLS model to mitigate drift effects, thereby allowing to extend the applicability of the model developed within an olfactometric campaign to the subsequent ones. The classification performance achieved involving the OPLS correction (i.e., 75%) was considerably higher than the one achieved on non-corrected data (i.e., about 55%).

Assessment of outdoor odor emissions from polluted sites using simultaneous thermal desorption-gas chromatography-mass spectrometry (TD-GC-MS), electronic nose in conjunction with advanced multivariate statistical approaches

Poor air quality, particularly in urban areas, causes various diseases and degrades living standards. Air quality could be affected by emissions of odor, Volatile Organic Compounds (VOCs), and other gases. Therefore, assessment and monitoring of odorous air quality using sensitive, simple, rapid, accurate and portable tools is very important for public health. This study aimed to characterize odor emissions to detect malfunctions in facilities and to prevent air pollution and olfactory nuisance in the environment. A gas chromatographic method, in conjunction with sensorial analysis were performed for odorous air samples analysis collected from neighborhood of Meknes city (Morocco). Advanced multivariate statistical approaches, such as Principal Components Analysis (PCA), Discriminant Function Analysis (DFA), Support Vector Machines (SVMs), and Hierarchical Cluster Analysis (HCA), were used to describe samples similarities. The electronic nose (e-nose) data processing exhibits a satisfactory discrimination between the odorous air samples. Twenty-four VOCs with known molecular formulas were identified with Thermal Desorption-Gas Chromatography-Mass Spectrometry (TD-GC-MS). A validated Partial Least Square (PLS) model foresees good calibration between e-nose measurement and TD-GC-MS analysis. The finding indicates that TD-GC–MS approach in conjunction with e-nose unit could be suitable tool for environmental measurement-based odor emissions.

A KECA identification method based on GA for E-nose data of six kinds of Chinese spirits

In order to improve the correct identification rate of six types of Chinese spirits using electronic nose (E-nose), the Kernel entropy component analysis (KECA) identification method combined with Genetic algorithm (GA) was proposed. Firstly, integral value (INV), relative steady-state average value (RSAV) and wavelet energy value (WEV) were extracted and employed to represent the E-nose data. Secondly, radial basis function (RBF) was selected as the kernel function, then the kernel parameter η of RBF was optimized by the matrix similarity measurement method and the GA. The corresponding optimized kernel parameter η was 16.8608 (matrix similarity measurement) and 67.9039 (GA), respectively. When the first 125 kernel entropy components were selected for Fisher discriminant analysis (FDA), the correct identification rate of FDA (KECA + FDA) combined with GA were 97.62 % and 98.81 % for the training set and testing set, respectively; the correct identification rate of FDA (KECA + FDA) combined with matrix similarity measurement were 93.58 and 91.67 % for the training set and testing set, respectively. Therefore, the kernel parameter η determined by GA was significantly better than that of matrix similarity measurement. Finally, the correct identification rate of FDA and KECA + FDA was compared, and the results of FDA were only 82.14 % and 79.92 % for the training set and testing set, respectively. The identification results of FDA were far worse than that of KECA + FDA. The KECA + FDA method combined with GA was suitable for the identification of the six types of Chinese spirits by E-nose.

Prediction of Sake Component Values Using E-nose and E-tongue Data by Machine Learning

Prediction of Sake Component Values Using E-nose and E-tongue Data by Machine Learning

Collaborative analysis on differences in volatile compounds of Harbin red sausages smoked with different types of woodchips based on gas chromatography–mass spectrometry combined with electronic nose

In this study, the volatile profiles of Harbin red sausages smoked with pear, oak, apple and beech woodchips were evaluated using the combination of solid-phase microextraction gas chromatography-mass spectrometry (SPME-GC/MS) and electronic nose (E-nose). Comparisons of the colour, texture profile and sensory attributes were also conducted. The E-nose data and GC/MS data were analysed by principal component analysis (PCA), respectively. The correlation between E-nose sensor responses and volatile compounds was evaluated with partial least squares regression (PLSR). The results showed that the smoking process played an important role in the formation of volatile compounds of Harbin red sausage. The smoking process caused an increase in the intensity of smoky odour, and the sausage smoked with different woodchips showed higher smoky odour compared with the unsmoked sausage (P < 0.05). A total of 87 volatile compounds were identified by GC/MS. According to the odour activity values, the characteristic odour of the Harbin red sausages mainly resulted from 22 volatile compounds. PCA of the volatile compounds and E-nose sensors could effectively differentiate sausages that were unsmoked and separately smoked with different woodchips. Furthermore, the PLSR results (P < 0.05, Q2 = 0.619) indicated that there was a high correlation between the E-nose sensors and volatile compounds.

Analysis of the volatile compounds’ condensate exhaled air “electronic nose” based on piezoelectric sensor to assess the status of calves

The article discusses general principles of obtaining diagnostic information using eight chemical piezoelectric gas sensors with nanostructural coverings from exhaled breath condensate for health assessment of the upper respiratory tract of pre-month-old calves. Multidimensional information of an e-nose can be presented in various numeric and visualized matrices, characteristics of which are an integral analytical signal of the sensor array. The research devoted to the search for the techniques of extracting analytical information from multidimensional e-nose data to assess upper respiratory tract state from the appearance of the first sign of respiratory disease to tracheobronchitis and bronchopneumonia. The piezoelectric sensor array is characterized by high sorption activity with priority biomolecules (which are the markers of the abnormal metabolic processes), low cost along with reliable repeatability of sorption properties from batch to batch, the simplicity of application, and fast response and recovery time. Here we present the results of simple algorithms used for assessment of upper respiratory tract state by a 2-minute analysis of odour over 1-ml biosample without sample preparation. It was shown that traditional quantitative parameters of e-nose are not adequate for simultaneous sample grouping and volatile compounds qualitative composition establishing. The additionally calculated sorption parameters Aij are more informative in the analysis of biosamples volatile: using for identification of volatile biomarkers, describe the health state correlating with clinical diagnosis. The sequence of information processing: signals of each sensor, integral characteristic, “visual prints,” additional sorption parameters–allows assessing the calf health virtually in situ without transporting samples to specialized laboratories.

A Comparison of Various Algorithms for Classification of Food Scents Measured with an Ion Mobility Spectrometry.

The present aim was to compare the accuracy of several algorithms in classifying data collected from food scent samples. Measurements using an electronic nose (eNose) can be used for classification of different scents. An eNose was used to measure scent samples from seven food scent sources, both from an open plate and a sealed jar. The k-Nearest Neighbour (k-NN) classifier provides reasonable accuracy under certain conditions and uses traditionally the Euclidean distance for measuring the similarity of samples. Therefore, it was used as a baseline distance metric for the k-NN in this paper. Its classification accuracy was compared with the accuracies of the k-NN with 66 alternative distance metrics. In addition, 18 other classifiers were tested with raw eNose data. For each classifier various parameter settings were tried and compared. Overall, 304 different classifier variations were tested, which differed from each other in at least one parameter value. The results showed that Quadratic Discriminant Analysis, MLPClassifier, C-Support Vector Classification (SVC), and several different single hidden layer Neural Networks yielded lower misclassification rates applied to the raw data than k-NN with Euclidean distance. Both MLP Classifiers and SVC yielded misclassification rates of less than when applied to raw data. Furthermore, when applied both to the raw data and the data preprocessed by principal component analysis that explained at least or of the total variance in the raw data, Quadratic Discriminant Analysis outperformed the other classifiers. The findings of this study can be used for further algorithm development. They can also be used, for example, to improve the estimation of storage times of fruit.

Characterization of flavor volatile compounds in industrial stir-frying mutton sao zi by GC-MS, E-nose, and physicochemical analysis.

This study aims to investigate the flavor changes of industrial stir‐frying mutton sao zi, a mutton product popular in the northwest of China, at different stir‐frying stages. Electronic nose (E‐nose) was used to recognize mutton sao zi odors at different processing time points, and the individual volatile compounds were further identified by the solid‐phase microextraction (SPME) combined with gas chromatography–mass spectrometry (GC‐MS). A total of 105 volatile compounds were detected by GC‐MS, of which 51 were major volatile compounds. Additionally, GC‐MS and E‐nose data of the samples were also correlated with the fatty acids, crude composition (moisture, fat, protein), and amino acids. The stir‐frying time and temperature may be the critical contributors to different flavors of industrial stir‐frying mutton sao zi. The signal intensities of W1S, W1W, W2S, W2W, and W3S sensors positively correlate with protein, fat, and 18 amino acids, but negatively with SFA and moisture. Hence, this study explored the flavor changes of industrial stir‐frying mutton sao zi by E‐nose and SPME‐GC‐MS for the first time, providing an insight into the industrial production and flavor control stir‐frying machine of stir‐frying mutton products with household flavor.