Qualitative and quantitative studies of phthalates in extra virgin olive oil (EVOO) by surface-enhanced Raman spectroscopy (SERS) combined with long short term memory (LSTM) neural network

Comparison of resmetirom quantative analysis in API and formulation models based on PXRD, FTIR and Raman scanning imaging combined with univariate and multivariate analyses

Resmetirom is the first innovative drug for the treatment of non-alcoholic steatohepatitis (NASH), and its two crystal forms, form I and form-2H2O, were mentioned in the original patent. The commercially available crystal form of resmetirom was form I. However, the tablets were subject to temperature, pressure, and humidity, and may be converted to form-2H2O, which affects the bioavailability and efficacy. Therefore, it is crucial to establish a suitable crystalline quantification method to examine the concentration of both crystalline forms in APIs and formulations. The main objective of this paper was to test the feasibility of PXRD, FTIR and Raman for quantitative analysis of form I content in API and formulation models. Commonly used methods to establish quantitative modelling were univariate and multivariate analyses, due to the overlapping peaks in the FTIR and Raman spectra of two forms, only the multivariate method, with partial least squares regression (PLSR), was used, both univariate and multivariate analysis were utilized in PXRD since it has distinct single peaks. In the multivariate models, the raw spectra are preprocessed to remove interfering information and spectral noise by nine commonly used pretreatment methods. According to the result of this study, all four calibration models could be applied to the quantitative analysis of form I in two models; however, Raman was found to be the most appropriate model for both API model (Y = 0.99523X+0.00300, R2 = 0.9997, RMSECV = 9.32 %, RESEP = 0.29 %, RESEC = 0.19 %, LOD = 3.2819 %, LOQ = 9.9451 %, MSC pretreated) and formulation model (Y = 0.99456X+0.00331, R2 = 0.9996, RMSECV = 1.14 %, RESEP = 0.39 %, RESEC = 0.01 %, LOD = 3.3098 %, LOQ = 9.1029 %, WT pretreated).

Quantitative determination of acid value in palm oil during thermal oxidation using Raman spectroscopy combined with deep learning models

Accurate monitoring of acid value (AV) is critical for edible oil quality control, yet traditional chemometric methods often face limitations in handling complex spectral data. This study combines Raman spectroscopy with deep learning, including Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Transformer, to explore their potential in improving the accuracy and efficiency of AV quantification during the thermal oxidation of palm oil. The results showed that all three deep learning models outperformed traditional chemometric methods in predictive accuracy. The CNN-LSTM model achieved the best performance, with a predicted coefficient of determination (Rp2) of 0.9978, a mean square error of prediction (RMSEP) of 0.0015, and a residual predictive deviation (RPD) of 21.21. This method demonstrates the effectiveness of Raman spectroscopy-driven deep learning for precise AV monitoring and holds promise for further validation with more diverse indicator datasets, providing a novel technical reference for edible oil quality control.

Multifunctional metal-organic frameworks-mediated colorimetric/photothermal immunosensor for highly sensitivity detection of dibutyl phthalate

Dibutyl phthalate (DBP), a priority pollutant among phthalic acid esters (PAEs) exhibits significant reproductive and respiratory toxicity. In this study, a multifunctional metal-organic frameworks-mediated colorimetric/photothermal immunosensor was established for the quantitative detection of DBP. Firstly, a highly sensitive and specific monoclonal antibody (mAb), designated 3A5, was prepared with a sensitivity IC50 value of 16.29 ng/mL. Secondly, a metal-organic framework material (ZrO₂@C) was synthesized via a two-step pyrolysis process of UiO-66. Subsequently, a multifunctional immunoprobe was prepared for the detection of DBP. Using ZrO₂@C-based colorimetric and photothermal dual-signal immunosensor ensured the accuracy of the detection results, as the multiple of these signals effectively guaranteed the reliability of the results. The limit of detection (LOD) for the dual signal was 0.766 ng/mL (colorimetric signal) and 0.465 ng/mL (photothermal signal). In conclusion, this work presented a novel and feasible approach for the development of multifunctional nanomaterials for the fabrication of immunosensors.

Advances in Food Aroma Analysis: Extraction, Separation, and Quantification Techniques

Decoding the aroma composition plays a key role in designing and producing foods that consumers prefer. Due to the complex matrix and diverse aroma compounds of foods, isolation and quantitative analytical methods were systematically reviewed. Selecting suitable and complementary aroma extraction methods based on their characteristics can provide more complete aroma composition information. Multiple mass spectrometry detectors (MS, MS/MS, TOF-MS, IMS) and specialized detectors, including flame ionization detector (FID), electron capture detector (ECD), nitrogen-phosphorus detector (NPD), and flame photometric detector (FPD), are the most important qualitative technologies in aroma identification and quantification. Furthermore, the real-time monitoring of aroma release and perception is an important developing trend in the aroma perception of future food. A combination of artificial intelligence for chromatographic analysis and characteristic databases could significantly improve the qualitative analysis efficiency and accuracy of aroma analysis. External standard method and stable isotope dilution analysis were the most popular quantification methods among the four quantification methods. The combination with flavoromics enables the decoding of aroma profile contributions and the identification of characteristic marker aroma compounds. Aroma analysis has a wide range of applications in the fields of raw materials selection, food processing monitoring, and products quality control.

Recent advances on artificial intelligence-based approaches for food adulteration and fraud detection in the food industry: Challenges and opportunities

Food adulteration is the deceitful practice of misleading consumers about food to profit from it. The threat to public health and food quality or nutritional valuable make it a major issue. Food origin and adulteration should be considered to safeguard customers against fraud. It has been established that artificial intelligence is a cutting-edge technology in food science and engineering. In this study, it has been explained how AI detects food tampering. Applications of AI such as machine learning tools in food quality have been studied. This review covered several food quality detection web-based information sources. The methods used to detect food adulteration and food quality standards have been highlighted. Various comparisons between state-of-the-art techniques, datasets, and outcomes have been conducted. The outcomes of this investigation will assist researchers choose the best food quality method. It will help them identify of foods that have been explored by researchers and potential research avenues.

Artificial Intelligence in Agro-Food Systems: From Farm to Fork

The current landscape of the food processing industry places a strong emphasis on improving food quality, nutritional value, and processing techniques. This focus arises from consumer demand for products that adhere to high standards of quality, sensory characteristics, and extended shelf life. The emergence of artificial intelligence (AI) and machine learning (ML) technologies is instrumental in addressing the challenges associated with variability in food processing. AI represents a promising interdisciplinary approach for enhancing performance across various sectors of the food industry. Significant advancements have been made to address challenges and facilitate growth within the food sector. This review highlights the applications of AI in agriculture and various sectors of the food industry, including bakery, beverage, dairy, food safety, fruit and vegetable industries, packaging and sorting, and the drying of fresh foods. Various strategies have been implemented across different food sectors to promote advancements in technology. Additionally, this article explores the potential for advancing 3D printing technology to enhance various aspects of the food industry, from manufacturing to service, while also outlining future perspectives.

Investigation of heat-induced pork batter quality detection and change mechanisms using Raman spectroscopy coupled with deep learning algorithms

Pork batter quality significantly affects its product. Herein, this study explored the use of Raman spectroscopy combined with deep learning algorithms for rapidly detecting pork batter quality and revealing the mechanisms of quality changes during heating. Results showed that heating increased β-sheet content (from 26.38 to 41.42%) and exposed hidden hydrophobic groups, which formed aggregates through chemical bonds. Dominant hydrophobic interactions further cross-linked these aggregates, establishing a more homogeneous and denser network at 80 °C. Subsequently, convolutional neural networks (CNN), long short-term memory neural networks (LSTM), and CNN-LSTM were comparatively used to predict gel strength and whiteness in batters based on the Raman spectrum. Thereinto, CNN-LSTM provided the optimal results for gel strength (Rp = 0.9515, RPD = 3.1513) and whiteness (Rp = 0.9383, RPD = 3.0152). Therefore, this study demonstrated the potential of Raman spectroscopy combined with deep learning algorithms as non-destructive tools for predicting pork batter quality and elucidating quality change mechanisms.

Machine Learning-Based Process Optimization in Biopolymer Manufacturing: A Review

The integration of machine learning (ML) into material manufacturing has driven advancements in optimizing biopolymer production processes. ML techniques, applied across various stages of biopolymer production, enable the analysis of complex data generated throughout production, identifying patterns and insights not easily observed through traditional methods. As sustainable alternatives to petrochemical-based plastics, biopolymers present unique challenges due to their reliance on variable bio-based feedstocks and complex processing conditions. This review systematically summarizes the current applications of ML techniques in biopolymer production, aiming to provide a comprehensive reference for future research while highlighting the potential of ML to enhance efficiency, reduce costs, and improve product quality. This review also shows the role of ML algorithms, including supervised, unsupervised, and deep learning algorithms, in optimizing biopolymer manufacturing processes.

Rapid quantitative analysis of soybean protein isolates secondary structure by two-dimensional correlation infrared spectroscopy through pH perturbation

The infrared spectroscopy (IR) signal of protein is prone to being covered by impurity signals, and the accuracy of the secondary structure content calculated using spectral data is poor. To tackle this challenge, a rapid high-precision quantitative model for protein secondary structure was proposed. Firstly, a two-dimensional correlation calculation was performed based on 60 groups of soybean protein isolates (SPI) infrared spectroscopy data, resulting in a two-dimensional correlation infrared spectroscopy (2DCOS-IR). Subsequently, the optimal characteristic bands of the four secondary structures were extracted from the 2DCOS-IR. Ultimately, partial least squares (PLS), long short-term memory (LSTM), and bidirectional long short-term memory (BILSTM) algorithms were used to model the extracted characteristic bands and predict the content of SPI secondary structure. The findings suggested that BILSTM combined with 2DCOS-IR model (2DCOS-BILSTM) exhibited superior predictive performance. The prediction sets for α-helix, β-sheet, β-turn, and random coil were designated as 0.9257, 0.9077, 0.9476, and 0.8443, respectively, and their corresponding RMSEP values were 0.26, 0.48, 0.20, and 0.15. This strategy enhances the precision of IR and facilitates the rapid identification of secondary structure components within SPI, which is vital for the advancement of protein industrial production.

Rapid and accurate quality evaluation of Angelicae Sinensis Radix based on near-infrared spectroscopy and Bayesian optimized LSTM network

The authentic traditional Chinese medicines (TCMs) including Angelicae Sinensis Radix (ASR) are the representative of high-quality herbals in China. However, ASR from authentic region being adulterated or counterfeited is frequently occurring, and there is still a lack of rapid quality evaluation methods for identifying the authentic ASR. In this study, the color features of ASR were firstly characterized. The results showed that the authentic ASR cannot be fully identified by color characteristics. Then near-infrared (NIR) spectroscopy combined with Bayesian optimized long short-term memory (BO-LSTM) was used to evaluate the quality of ASR, and the performance of BO-LSTM with common classification and regression algorithms was compared. The results revealed that following the pretreatment of NIR spectra, the optimal NIR spectra combined with BO-LSTM not only successfully distinguished authentic, non-authentic, and adulterated ASR with 100 % accuracy, but also accurately predicted the adulteration concentration of authentic ASR (R2 > 0.99). Moreover, BO-LSTM demonstrated excellent performance in classification and regression compared with common algorithms (ANN, SVM, PLSR, etc.). Overall, the proposed strategy could quickly and accurately evaluate the quality of ASR, which provided a reference for other TCMs.

Nanotechnology-based optical sensors for Baijiu quality and safety control

Baijiu quality and safety have received considerable attention owing to the gradual increase in its consumption. However, owing to the unique and complex process of Baijiu production, issues leading to quality and safety concerns may occur during the manufacturing process. Therefore, establishing appropriate analytical methods is necessary for Baijiu quality assurance and process control. Nanomaterial (NM)-based optical sensing techniques have garnered widespread interest because of their unique advantages. However, comprehensive studies on nano-optical sensing technology for quality and safety control of Baijiu are lacking. In this review, we systematically summarize NM-based optical sensor applications for the accurate detection and quantification of analytes closely related to Baijiu quality and safety. Furthermore, we evaluate the sensing mechanisms for each application. Finally, we discuss the challenges nanotechnology poses for Baijiu analysis and future trends. Overall, nanotechnological approaches provide a potentially useful alternative for simplifying Baijiu analysis and improving final product quality and safety.

Recent advances in artificial intelligence towards the sustainable future of agri-food industry

Artificial intelligence has the potential to alter the agricultural and food processing industries, with significant ramifications for sustainability and global food security. The integration of artificial intelligence in agriculture has witnessed a significant uptick in recent years. Therefore, comprehensive understanding of these techniques is needed to broaden its application in agri-food supply chain. In this review, we explored cutting-edge artificial intelligence methodologies with a focus on machine learning, neural networks, and deep learning. The application of artificial intelligence in agri-food industry and their quality assurance throughout the production process is thoroughly discussed with an emphasis on the current scientific knowledge and future perspective. Artificial intelligence has played a significant role in transforming agri-food systems by enhancing efficiency, sustainability, and productivity. Many food industries are implementing the artificial intelligence in modelling, prediction, control tool, sensory evaluation, quality control, and tackling complicated challenges in food processing. Similarly, artificial intelligence applied in agriculture to improve the entire farming process, such as crop yield optimization, use of herbicides, weeds identification, and harvesting of fruits. In summary, the integration of artificial intelligence in agri-food systems offers the potential to address key challenges in agriculture, enhance sustainability, and contribute to global food security.

Rapid Indentification of Auramine O Dyeing Adulteration in Dendrobium officinale, Saffron and Curcuma by SERS Raman Spectroscopy Combined with SSA-BP Neural Networks Model

(1) Background: Rapid and accurate determination of the content of the chemical dye Auramine O(AO) in traditional Chinese medicines (TCMs) is critical for controlling the quality of TCMs. (2) Methods: Firstly, various models were developed to detect AO content in Dendrobium officinale (D. officinale). Then, the detection of AO content in Saffron and Curcuma using the D. officinale training set as a calibration model. Finally, Saffron and Curcuma samples were added to the training set of D. officinale to predict the AO content in Saffron and Curcuma using secondary wavelength screening. (3) Results: The results show that the sparrow search algorithm (SSA)-backpropagation (BP) neural network (SSA-BP) model can accurately predict AO content in D. officinale, with Rp2 = 0.962, and RMSEP = 0.080 mg/mL. Some Curcuma samples and Saffron samples were added to the training set and after the secondary feature wavelength screening: The Support Vector Machines (SVM) quantitative model predicted Rp2 fluctuated in the range of 0.780 ± 0.035 for the content of AO in Saffron when 579, 781, 1195, 1363, 1440, 1553 and 1657 cm-1 were selected as characteristic wavelengths; the Partial Least Squares Regression (PLSR) model predicted Rp2 fluctuated in the range of 0.500 ± 0.035 for the content of AO in Curcuma when 579, 811, 1195, 1353, 1440, 1553 and 1635 cm-1 were selected as the characteristic wavelengths. The robustness and generalization performance of the model were improved. (4) Conclusion: In this study, it has been discovered that the combination of surface-enhanced Raman spectroscopy (SERS) and machine learning algorithms can effectively and promptly detect the content of AO in various types of TCMs.