Identification and quantification of adulterated honey by Raman spectroscopy combined with convolutional neural network and chemometrics

Detecting Honey Adulteration: Advanced Approach Using UF-GC Coupled with Machine Learning

This article introduces a novel approach to detecting honey adulteration by combining ultra-fast gas chromatography (UF-GC) with advanced machine learning techniques. Machine learning models, particularly support vector regression (SVR) and least absolute shrinkage and selection operator (LASSO), were applied to predict adulteration in orange blossom (OB) and sunflower (SF) honeys. The SVR model achieved R2 values above 0.90 for combined honey types. Treating OB and SF honeys separately resulted in a significant accuracy improvement, with R2 values exceeding 0.99. LASSO proved especially effective when honey types were treated individually. The integration of UF-GC with machine learning not only provides a reliable method for detecting honey adulteration, but also sets a precedent for future research in the application of this technique to other food products, potentially enhancing food authenticity across the industry.

Application of Spatial Offset Raman Spectroscopy (SORS) and Machine Learning for Sugar Syrup Adulteration Detection in UK Honey

Honey authentication is a complex process which traditionally requires costly and time-consuming analytical techniques not readily available to the producers. This study aimed to develop non-invasive sensor methods coupled with a multivariate data analysis to detect the type and percentage of exogenous sugar adulteration in UK honeys. Through-container spatial offset Raman spectroscopy (SORS) was employed on 17 different types of natural honeys produced in the UK over a season. These samples were then spiked with rice and sugar beet syrups at the levels of 10%, 20%, 30%, and 50% w/w. The data acquired were used to construct prediction models for 14 types of honey with similar Raman fingerprints using different algorithms, namely PLS-DA, XGBoost, and Random Forest, with the aim to detect the level of adulteration per type of sugar syrup. The best-performing algorithm for classification was Random Forest, with only 1% of the pure honeys misclassified as adulterated and <3.5% of adulterated honey samples misclassified as pure. Random Forest was further employed to create a classification model which successfully classified samples according to the type of adulterant (rice or sugar beet) and the adulteration level. In addition, SORS spectra were collected from 27 samples of heather honey (24 Calluna vulgaris and 3 Erica cinerea) produced in the UK and corresponding subsamples spiked with high fructose sugar cane syrup, and an exploratory data analysis with PCA and a classification with Random Forest were performed, both showing clear separation between the pure and adulterated samples at medium (40%) and high (60%) adulteration levels and a 90% success at low adulteration levels (20%). The results of this study demonstrate the potential of SORS in combination with machine learning to be applied for the authentication of honey samples and the detection of exogenous sugars in the form of sugar syrups. A major advantage of the SORS technique is that it is a rapid, non-invasive method deployable in the field with potential application at all stages of the supply chain.

Portable Raman spectroscopy coupled with PLSR analysis for monitoring and predicting of the quality of fresh-cut Chinese yam at different storage temperatures

Portable Raman spectroscopy coupled with partial least squares regression (PLSR) model was performed for monitoring and predicting four quality indicators, moisture content, water activity, polysaccharide content and microbial content of the fresh-cut Chinese yam at different storage temperatures. The variations in the four key indicators were first depicted through a spider web diagram as the storage temperature changed. More importantly, the four key indicators can be accurately monitored and predicted through optimized PLSR models combining with Raman spectroscopy. Among all of the PLSR models for the four indicators, the regression model for moisture content was relatively the best. In addition, storage temperature played a significant role on the model performance of PLSR. The model performance for all indicators at room temperature and high temperature was better than the corresponding PLSR models at refrigeration and freezing conditions. Especially at 25 ℃, the R2 in the calibration set basically reached 0.9. These observations indicated that portable Raman spectroscopy, a simple and easy-to-use detection technique, can monitor and predict the multiple quality indicators of fresh-cut Chinese yam combined with effectively PLSR model, which would be conducive to their applications in food industry.

Spectral encoder to extract the efficient features of Raman spectra for reliable and precise quantitative analysis

Raman spectroscopy has become a powerful analytical tool highly demanded in many applications such as microorganism sample analysis, food quality control, environmental science, and pharmaceutical analysis, owing to its non-invasiveness, simplicity, rapidity and ease of use. Among them, quantitative research using Raman spectroscopy is a crucial application field of spectral analysis. However, the entire process of quantitative modeling largely relies on the extraction of effective spectral features, particularly for measurements on complex samples or in environments with poor spectral signal quality. In this paper, we propose a method of utilizing a spectral encoder to extract effective spectral features, which can significantly enhance the reliability and precision of quantitative analysis. We built a latent encoded feature regression model; in the process of utilizing the autoencoder for reconstructing the spectrometer output, the latent feature obtained from the intermediate bottleneck layer is extracted. Then, these latent features are fed into a deep regression model for component concentration prediction. Through detailed ablation and comparative experiments, our proposed model demonstrates superior performance to common methods on single-component and multi-component mixture datasets, remarkably improving regression precision while without needing user-selected parameters and eliminating the interference of irrelevant and redundant information. Furthermore, in-depth analysis reveals that latent encoded feature possesses strong nonlinear feature representation capabilities, low computational costs, wide adaptability, and robustness against noise interference. This highlights its effectiveness in spectral regression tasks and indicates its potential in other application fields. Sufficient experimental results show that our proposed method provides a novel and effective feature extraction approach for spectral analysis, which is simple, suitable for various methods, and can meet the measurement needs of different real-world scenarios.

Raman spectroscopy combined with multiple one-dimensional deep learning models for simultaneous quantification of multiple components in blended olive oil

Blended vegetable oils are highly prized by consumers for their comprehensive nutritional profile. Therefore, there is an urgent need for a rapid and accurate method to identify the true content of blended oils. This study combined Raman spectroscopy with three deep learning models (CNN-LSTM, improved AlexNet, and ResNet) to simultaneously quantify extra virgin olive oil (EVOO), soybean oil, and sunflower oil in olive blended oil. The results demonstrate that all three deep learning models exhibited superior predictive ability compared to traditional chemometric methods. Specifically, the CNN-LSTM model achieved a coefficient of determination (R2p) of over 0.995 for each oil in the quantitative analysis of three-component blended oils, with a mean square error of prediction (RMSEP) of less than 2%. This study presents a novel approach for the simultaneous quantitative analysis of multi-component blended oils, providing a rapid and accurate method for the identification of falsely labeled blended oils.

Non-Targeted Detection and Quantification of Food Adulteration of High-Quality Stingless Bee Honey (SBH) via a Portable LED-Based Fluorescence Spectroscopy

Stingless bee honey (SBH) is rich in phenolic compounds and available in limited quantities. Authentication of SBH is important to protect SBH from adulteration and retain the reputation and sustainability of SBH production. In this research, we use portable LED-based fluorescence spectroscopy to generate and measure the fluorescence intensity of pure SBH and adulterated samples. The spectrometer is equipped with four UV-LED lamps (peaking at 365 nm) as an excitation source. Heterotrigona itama, a popular SBH, was used as a sample. 100 samples of pure SBH and 240 samples of adulterated SBH (levels of adulteration ranging from 10 to 60%) were prepared. Fluorescence spectral acquisition was measured for both the pure and adulterated SBH samples. Principal component analysis (PCA) demonstrated that a clear separation between the pure and adulterated SBH samples could be established from the first two principal components (PCs). A supervised classification based on soft independent modeling of class analogy (SIMCA) achieved an excellent classification result with 100% accuracy, sensitivity, specificity, and precision. Principal component regression (PCR) was superior to partial least squares regression (PLSR) and multiple linear regression (MLR) methods, with a coefficient of determination in prediction (R2p) = 0.9627, root mean squared error of prediction (RMSEP) = 4.1579%, ratio prediction to deviation (RPD) = 5.36, and range error ratio (RER) = 14.81. The LOD and LOQ obtained were higher compared to several previous studies. However, most predicted samples were very close to the regression line, which indicates that the developed PLSR, PCR, and MLR models could be used to detect HFCS adulteration of pure SBH samples. These results showed the proposed portable LED-based fluorescence spectroscopy has a high potential to detect and quantify food adulteration in SBH, with the additional advantages of being an accurate, affordable, and fast measurement with minimum sample preparation.

On-line multi-gas component measurement in the mud logging process based on Raman spectroscopy combined with a CNN-LSTM-AM hybrid model

The qualitative and quantitative analysis of gas components extracted from drilling fluids during mud logging is essential for identifying drilling anomalies, reservoir characteristics, and hydrocarbon properties during oilfield recovery. Gas chromatography (GC) and gas mass spectrometers (GMS) are currently used for the online analysis of gases throughout the mud logging process. Nevertheless, these methods have limitations, including expensive equipment, high maintenance costs, and lengthy detection periods. Raman spectroscopy can be applied to the online quantification of gases at mud logging sites due to its in-situ analysis, high resolution, and rapid detection. However, laser power fluctuations, field vibrations, and the overlapping of characteristic peaks of different gases in the existing online detection system of Raman spectroscopy can affect the quantitative accuracy of the model. For these reasons, a gas Raman spectroscopy system with a high reliability, low detection limits, and increased sensitivity has been designed and applied to the online quantification of gases in the mud logging process. The near-concentric cavity structure is used to improve the signal acquisition module in the gas Raman spectroscopic system, thus enhancing the Raman spectral signal of the gases. One-dimensional convolutional neural networks (1D-CNN) combined with long- and short-term memory networks (LSTM) are applied to construct quantitative models based on the continuous acquisition of Raman spectra of gas mixtures. In addition, the attention mechanism is used to futher improve the quantitative model performance. The results indicated that our proposed method has the capability to continuously on-line detect 10 hydrocarbon and non-hydrocarbon gases in the mud logging process. The limitation of detection (LOD) for different gas components based on the proposed method are in the range of 0.0035%-0.0223%. Based on the proposed CNN-LSTM-AM model, the average detection errors of different gas components range from 0.899% to 3.521%, and their maximum detection errors range from 2.532% to 11.922%. These results demonstrate that our proposed method has a high accuracy, low deviation, and good stability and can be applied to the on-line gas analysis process in the mud logging field.

An extreme learning machine optimized by differential evolution and artificial bee colony for predicting the concentration of whole blood with Fourier Transform Raman spectroscopy

Raman spectroscopy, with its advantages of non-contact nature, rapid detection, and minimum water interference, is promising for non-invasive blood detection or diagnosis in clinic applications. However, there is a critical issue that how to accurately analyze blood composition by Raman spectroscopy. In this study, we apply extreme learning machine (ELM) algorithm and a multivariate calibration regression model to analyze the results from Raman spectroscopy and determine the component's concentrations in blood samples, including glucose, cholesterol, and triglyceride. Self-adaption differential evolution artificial bee colony (SADEABC) algorithm was further applied to increase the data's accuracy and robustness. The obtained data for coefficient of determination, root mean square error of calibration, root mean square error of prediction, and relative percent deviation, were 0.9822, 0.3993, 0.3827, and 6.6679 for glucose, 0.9786, 0.2104, 0.2088 and 5.9533 for cholesterol, and 0.9921, 0.2744, 0.3433 and 10.5075 for triglyceride, respectively. Results demonstrated that the model based on SADEABC-ELM show much better prediction data than those models based on the ELM and ABC-ELM.

Determination of sugar content in honey using LC-Raman and programmable pump-Raman methods

We combined (i) liquid chromatography and Raman spectrometry (LC-Raman) and (ii) programmable pump and Raman spectrometry (PP-Raman) to separate and identify compounds in a mixture. These techniques were applied to conduct a quantitative analysis of the sugars in honey. The spectral and temporal axes of the LC-Raman data were analyzed using the MCR-ALS analysis procedure, which enabled the separation and identification of four sugars (glucose, fructose, sucrose, and trehalose). The PP-Raman method was employed to examine the sugar concentration dependence of the intensity pattern of the Raman spectrum, and the linear concentration dependence of the intensity was obtained. The sugar contents were quantitatively determined from the integrated area of the elution peaks. The result was consistent with those derived from mass spectrometry and previous studies. The origin of the errors in the derived sugar contents is discussed. Our study presents a novel quantitative LC-Raman spectrometric method that does not rely on resonance or surface enhancement effects.

Accurate identification of living Bacillus spores using laser tweezers Raman spectroscopy and deep learning

Accurately, rapidly, and noninvasively identifying Bacillus spores can greatly contribute to controlling a plenty of infectious diseases. Laser tweezers Raman spectroscopy (LTRS) has confirmed to be a powerful tool for studying Bacillus spores at a single cell level. In this study, we constructed a single-cell Raman spectra dataset of living Bacillus spores and utilized deep learning approach to accurately, nondestructively identify Bacillus spores. The trained convolutional neural network (CNN) could efficiently extract tiny Raman spectra features of five spore species, and provide a prediction accuracy of specie identification as high as 100 %. Moreover, the spectral feature differences in three Raman bands at 660, 826, and 1017 cm^-1 were confirmed to mostly contribute to producing such high prediction accuracy. In addition, optimal CNN model was employed to monitor and identify sporulation process at different metabolic phases in one growth cycle. The obtained average prediction accuracy of metabolic phase identification was approximately 88 %. It can be foreseen that, LTRS combined with CNN approach have great potential for accurately identifying spore species and metabolic phases at a single cell level, and can be gradually extended to perform identification for many unculturable bacteria growing in soil, water, and food.

Identification of antibiotic residues in aquatic products with surface-enhanced Raman scattering powered by 1-D convolutional neural networks

Universal and fast antibiotic residues detection technology is imperative for the control of food safety in aquatic products. However, accurate surface-enhanced Raman scattering (SERS) quantitative detection of complicated samples is still a challenge. A recognition method powered by deep learning and took advantage of the unique fingerprint information merits of SERS was proposed. Herein, the spectra were collected by Ag nanofilm SERS substrate prepared by self-assembly of Ag nanoparticles on water/oil interface. A SERS-based database of commonly used antibiotics in aquatic products was set up, which is suitable for employed as input data for learning and training. The results show that the five types of antibiotics are successfully distinguished through principal component analysis (PCA) and each antibiotic in every type was successfully distinguished. Furthermore, one-dimensional convolutional neural networks (1-D CNN) was used to distinguish the antibiotics, and the results show that all the test samples were correctly predicted by 1-D CNN model. The results of this research suggest the great potential of the combination of SERS spectra with deep learning as a method for rapid and highly accurate identification of antibiotic residues in aquatic products.

Insight into the Recent Application of Chemometrics in Quality Analysis and Characterization of Bee Honey during Processing and Storage

The application of chemometrics, a widely used science in food studies (and not only food studies) has begun to increase in importance with chemometrics being a very powerful tool in analyzing large numbers of results. In the case of honey, chemometrics is usually used for assessing honey authenticity and quality control, combined with well-established analytical methods. Research related to investigation of the quality changes in honey due to modifications after processing and storage is rare, with a visibly increasing tendency in the last decade (and concentrated on investigating novel methods to preserve the honey quality, such as ultrasound or high-pressure treatment). This review presents the evolution in the last few years in using chemometrics in analyzing honey quality during processing and storage. The advantages of using chemometrics in assessing honey quality during storage and processing are presented, together with the main characteristics of some well-known chemometric methods. Chemometrics prove to be a successful tool to differentiate honey samples based on changes of characteristics during storage and processing.

A deep learning model designed for Raman spectroscopy with a novel hyperparameter optimization method

Raman spectroscopy is a spectroscopic technique typically used to determine vibrational modes of molecules and is commonly used in chemistry to provide a structural fingerprint by which molecules can be identified. With the help of deep learning, Raman spectroscopy can be analyzed more efficiently and thus provide more accurate molecular information. However, no general neural network is designed for one-dimensional Raman spectral data so far. Furthermore, different combinations of hyperparameters of neural networks lead to results with significant differences, so the optimization of hyperparameters is a crucial issue in deep learning modeling. In this work, we propose a deep learning model designed for Raman spectral data and a hyperparameter optimization method to achieve its best performance, i.e., a method based on the simulated annealing algorithm to optimize the hyperparameters of the model. The proposed model and optimization method have been fully validated in a glioma Raman spectroscopy dataset. Compared with other published methods including linear regression, support vector regression, long short-term memory, VGG and ResNet, the mean squared error is reduced by 0.1557 while the coefficient determination is increased by 0.1195 on average.

Innovative Application of SERS in Food Quality and Safety: A Brief Review of Recent Trends

Innovative application of surface-enhanced Raman scattering (SERS) for rapid and nondestructive analyses has been gaining increasing attention for food safety and quality. SERS is based on inelastic scattering enhancement from molecules located near nanostructured metallic surfaces and has many advantages, including ultrasensitive detection and simple protocols. Current SERS-based quality analysis contains composition and structural information that can be used to establish an electronic file of the food samples for subsequent reference and traceability. SERS is a promising technique for the detection of chemical, biological, and harmful metal contaminants, as well as for food poisoning, and allergen identification using label-free or label-based methods, based on metals and semiconductors as substrates. Recognition elements, including immunosensors, aptasensors, or molecularly imprinted polymers, can be linked to SERS tags to specifically identify targeted contaminants and perform authenticity analysis. Herein, we highlight recent studies on SERS-based quality and safety analysis for different foods categories spanning the whole food chain, ‘from farm to table’ and processing, genetically modified food, and novel foods. Moreover, SERS detection is a potential tool that ensures food safety in an easy, rapid, reliable, and nondestructive manner during the COVID-19 pandemic.