Exploration of total synchronous fluorescence spectroscopy combined with pre-trained convolutional neural network in the identification and quantification of vegetable oil

Data integrity of food and machine learning: Strategies, advances and prospective

Data integrity is an emerging concept aimed at recording real food properties in the form of data throughout the food lifecycle. However, due to the one-sided nature of current food control data, the comprehensive implementation of data integrity has not been fully achieved. Cause food data integrity realization is required to establish the connection of data-algorithm-application. Machine learning (ML) provides a possibility for the practical carrier of food data integrity. Despite ML is one of top-trend in food quality and safety, ML applications are floating on the surface. The current review does not reveal the relationships behind different algorithms and data patterns. Similarly, due to the rapid development of ML, the current advanced concepts and data explanation tools have not been systematically reviewed. This paper expounds the feasibility of machine learning to achieve data integrity and looks forward to the future vision brought about by artificial intelligence to data integrity.

Chemical characterization and classification of vegetable oils using DESI-MS coupled with a neural network

This study tackled mislabeling fraud in vegetable oils, driven by price disparities and profit motives, by developing an approach combining desorption electrospray ionization mass spectrometry (DESI-MS) with a shallow convolutional neural network (SCNN). The method was designed to characterize lipids and distinguish between nine vegetable oils: corn, soybean, peanut, sesame, rice bran, sunflower, camellia, olive, and walnut oils. The optimized DESI-MS method enhanced the ionization of non-polar glycerides and detected ion adducts like [TG + Na]+, [TG + NH4]+. This process identified 53 lipid peaks, forming a robust lipid fingerprint for each oil type. An SCNN model was developed using fingerprints, achieving an impressive classification accuracy of 98.5 ± 2.2 %. The integration of DESI-MS with SCNN provides a fast and reliable tool for identifying and classifying vegetable oils, thereby reducing mislabeling fraud and assuring oil quality. By enabling accurate authentication, it contributes to improved transparency and integrity in food labeling and quality control practices.

A crude, cold-pressed oil from elderberry (Sambucus nigra L.) seeds: Comprehensive approach to properties and characterization using HPLC, DSC, and multispectroscopic methods

The physicochemical characterization of fresh, undiluted, cold-pressed oil from elderberry seeds (EO) is presented. The results showed EO's uniqueness for the 93 % presence of essential fatty acids, including linoleic n-6 (41 %), α-linolenic n-3 (38 %), and oleic n-9 (13 %) acids with favorable ratios for human nutrition, n-3/n-6 = 0.93. A γ-tocopherol is the dominant tocopherol (96 %), with a concentration of 20.62 mg/100 g, indicating low oil oxidative stability. DSC heating and cooling traces determined the thermal properties. These results also revealed the presence of metastable triacylglycerol (TAG) structures composed of polyunsaturated fatty acids. The presence of characteristic groups for fatty acids and TAGs in EO was confirmed by FTIR-ATR spectra. For the first time, Langmuir monolayer studies on EO revealed its low compressibility, indicating its low emulsifiability, and the presence of minor components of EO, including tocopherols, phenolic acids, polyphenols, flavonoids, and carotenoids, was determined using UV-Vis absorption and fluorescence excitation-emission matrix (EEM) along with the chemometric method.

Research Progress of Machine Learning in Extending and Regulating the Shelf Life of Fruits and Vegetables

Fruits and vegetables are valued for their flavor and high nutritional content, but their perishability and seasonality present challenges for storage and marketing. To address these, it is essential to accurately monitor their quality and predict shelf life. Unlike traditional methods, machine learning efficiently handles large datasets, identifies complex patterns, and builds predictive models to estimate food shelf life. These models can be continuously refined with new data, improving accuracy and robustness over time. This article discusses key machine learning methods for predicting shelf life and quality control of fruits and vegetables, with a focus on storage conditions, physicochemical properties, and non-destructive testing. It emphasizes advances such as dataset expansion, model optimization, multi-model fusion, and integration of deep learning and non-destructive testing. These developments aim to reduce resource waste, provide theoretical basis and technical guidance for the formation of modern intelligent agricultural supply chains, promote sustainable green development of the food industry, and foster interdisciplinary integration in the field of artificial intelligence.

Recent advances in shelf life prediction models for monitoring food quality

Each year, 1.3 billion tons of food is lost due to spoilage or loss in the supply chain, accounting for approximately one third of global food production. This requires a manufacturer to provide accurate information on the shelf life of the food in each stage. Various models for monitoring food quality have been developed and applied to predict food shelf life. This review classified shelf life models and detailed the application background and characteristics of commonly used models to better understand the different uses and aspects of the commonly used models. In particular, the structural framework, application mechanisms, and numerical relationships of commonly used models were elaborated. In addition, the study focused on the application of commonly used models in the food field. Besides predicting the freshness index and remaining shelf life of food, the study addressed aspects such as food classification (maturity and damage) and content prediction. Finally, further promotion of shelf life models in the food field, use of multivariate analysis methods, and development of new models were foreseen. More reliable transportation, processing, and packaging methods could be screened out based on real-time food quality monitoring.

Implementing neural network approach to create carbon-based optical nanosensor of heavy metal ions in liquid media

The present study is devoted to the creation of multifunctional optical carbon dots-based nanosensor to simultaneously measure concentrations of metal ions (Cu²⁺, Ni²⁺, Cr³⁺) and NO₃^- anion in liquid media. Such nanosensor operates on the basis of its fluorescence (FL) change under the influence of ions in the medium. However, the absence of analytical model, describing CD FL mechanism, the superposition of various luminescence quenching mechanisms during the interaction of carbon dots (CD) with cations, hampers the usage of classical approaches to solve this inverse multiparametric spectroscopic problem. To solve it neural networks were used that analyzed complex fluorescence signal from CD aqueous suspensions comprising Cu²⁺, Ni²⁺, Cr³⁺, NO₃^- ions in the concentration range from 0 to 4.95 mM. The following neural network architectures ensured optical spectroscopy inverse problem solution: multilayer perceptrons, 1D and 2D convolutional neural networks. The developed sensor enables simultaneous determination of the concentrations of heavy metal ions Cu²⁺, Ni²⁺, Cr³⁺ with a root mean squared error of 0.28 mM, 0.79 mM, 0.24 mM respectively. Based on the data given in the literature we can assert that the accuracy of the studied nanosensor satisfies the needs of monitoring the composition of waste and technological water. The developed nanosensor has a unique multimodality: with the simplicity of the synthesis protocol the sensor enables simultaneous determination of three heavy metal ions concentrations, while analogues are being developed mainly to measure the concentration of one (in rare cases two) heavy metal ions.

Olive oil classification with Laser-induced fluorescence (LIF) spectra using 1-dimensional convolutional neural network and dual convolution structure model

Laser-induced fluorescence (LIF) spectroscopy is widely used for the analysis and classification of olive oil. This paper proposes the classification of LIF data using a specific 1-dimensional convolutional neural network (1D-CNN) model, which does not require pre-processing steps such as normalisation or denoising and can be flexibly applied to massive data. However, by adding a dual convolution structure (Dual-conv) to the model, the features of the 1-dimensional spectra are more scattered within one convolution-pooling process; thus, the classification effects are improved. The models were validated through an olive oil classification experiment which contained a total of 72,000 sets of LIF spectra data, and the classification accuracy rate reached ∼99.69%. Additionally, a common classification approach, the support vector machine (SVM), was utilised for the comparison of the results. The results show that the neural networks perform better than the SVM. The Dual-conv model structure has a faster convergence speed and higher evaluation parameters than those of the 1D-CNN in the same period of iterations, without increasing the data dimension.

Quantitative analysis of blended corn-olive oil based on Raman spectroscopy and one-dimensional convolutional neural network

Blended vegetable oil is a vital product in the vegetable oil market, and quantifying high-value vegetable oil is of great significance to protect the rights and interests of consumers. In this study, we established a one-dimensional convolutional neural network (1D CNN) quantitative identification model based on Raman spectra to identify the amount of olive oil in a corn-olive oil blend. The results show that the 1D CNN model based on 315 extended average Raman spectra can quantitatively identify the content of olive oil, with R²p and RMSEP values of 0.9908 and 0.7183 respectively. Compared with partial least squares regression (PLSR) and support vector regression (SVR), although the index is not optimal, it provides a new analytical method for the quantitative identification of vegetable oil.

Deep learning aided quantitative analysis of anti-tuberculosis fixed-dose combinatorial formulation by terahertz spectroscopy

Anti-tuberculosis fixed-dose combinatorial formulation (FDCs) is an effective drug for the treatment of tuberculosis. As a compound medicine, its efficacy is based on the comprehensive action of multiple main ingredients. If the content of an active ingredient is insufficient, not only will it reduce the curative effect, but it also causes patients to develop drug resistance and leads to the evolution of drug-resistant strains of tuberculosis, which hamper the treatment of the disease. Thus accurate detection of the contents of active components in the anti-tuberculosis FDC is the key link of its quality control. For the first time, convolutional neural networks (CNN), one of the most popular deep learning methods, is adopted to develop a quantitative calibration model based on terahertz time-domain spectroscopy (THz-TDS) for the accurate detection of the content of each active component in the anti-tuberculosis FDCs. For comparison with CNN, partial least squares regression (PLSR) was also introduced to build a reference quantitative calibration model. For CNN modeling, the raw THz spectral is fed to the model directly; While for PLSR, prior to the spectrum feeding to the model, the raw spectral data are processed by multiple different combinations of preprocessing. Experimental and simulation results demonstrate that although preprocessing techniques can improve the prediction performance of PLSR, its prediction capabilities is still inferior to CNN based on raw spectrum. Therefore, for the quantitative analysis of the content of each active component in the anti-tuberculosis FDCs, CNN appears to be an ideal modeling method.