Reproducible, shelf-stable, and bioaffinity SERS nanotags inspired by multivariate polyphenolic chemistry for bacterial identification

"One-Pot" Readout Cyano-Programmable SERS-Encoded Platform Enables Ultrasensitive and Interference-Free Detection of Multitarget Bioamines

Surface-enhanced Raman spectroscopy (SERS) detection platforms with high signal-to-noise ratio in the "biological-silent" region (1800-2800 cm-1) are presently being developed for sensing and imaging applications, overcoming the limitations of traditional SERS studies in the "fingerprint" region. Herein, a series of cyano-programmable Raman reporters (RRs) operating in the "biological-silent" region were designed based on 4-mercaptobenzonitrile derivatives and then embedded in core-shell Au@Ag nanostars using a "bottom-up" strategy to provide SERS enhancement and encapsulation protection. The approach enabled the "one-pot" readout interference-free detection of multiple bioamines (histamine, tyramine, and β-phenethylamine) based on aptamer-driven magnetic-induced technology. Three cyano-encoded SERS tags resulted in separate SERS signals for histamine, tyramine, and β-phenethylamine at 2220, 2251, and 2150 cm-1, respectively. A target-specific aptamer-complementary DNA competitive binding strategy allowed the formation of microscale core-satellite assemblies between Fe3O4-based magnetic beads and the SERS tags, enabling multiple SERS signals to be observed simultaneously under a 785 nm laser excitation laser. The LODs for detection of the three bioamines were 0.61 × 10-5, 2.67 × 10-5, and 1.78 × 10-5 mg L-1, respectively. The SERS-encoded platform utilizing programmable reporters provides a fast and sensitive approach for the simultaneous detection of multiple biomarkers, paving the way for routine SERS analyses of multiple analytes in complex matrices.

A tailored dual core-shell magnetic SERS substrate with precise shell-thickness control for trace organophosphorus pesticides residues detection

For addressing the challenges of strong affinity SERS substrate to organophosphorus pesticides (OPs), herein, a rapid water-assisted layer-by-layer heteronuclear growth method was investigated to grow uniform UiO-66 shell with controllable thickness outside the magnetic core and provide abundant defect sites for OPs adsorption. By further assembling the tailored Au@Ag, a highly sensitive SERS substrate Fe3O4-COOH@UiO-66/Au@Ag (FCUAA) was synthesized with a SERS enhancement factor of 2.11 × 107. The substrate's suitability for the actual vegetable samples (cowpeas and peppers) was confirmed under both destructive and non-destructive detection conditions, showing a strong SERS response to fenthion and triazophos, with limits of detection of 1.21 × 10-5 and 2.96 × 10-3 mg/kg in the vegetables under destructive conditions, and 0.13 and 1.39 ng/cm2 for non-destructive detection, respectively. The FCUAA substrate had high SERS performance, effective adsorption capability for OPs, and demonstrated good applicability, thus exhibiting great potential for rapid detection of trace OPs residues in the food industry.

Fast real-time monitoring of meat freshness based on fluorescent sensing array and deep learning: From development to deployment

A fluorescent sensor array (FSA) combined with deep learning (DL) techniques was developed for meat freshness real-time monitoring from development to deployment. The array was made up of copper metal nanoclusters (CuNCs) and fluorescent dyes, having a good ability in the quantitative and qualitative detection of ammonia, dimethylamine, and trimethylamine gases with a low limit of detection (as low as 131.56 ppb) in range of 5 ∼ 1000 ppm and visually monitoring the freshness of various meats stored at 4 °C. Moreover, SqueezeNet was applied to automatically identify the fresh level of meat based on FSA images with high accuracy (98.17 %) and further deployed in various production environments such as personal computers, mobile devices, and websites by using open neural network exchange (ONNX) technique. The entire meat freshness recognition process only takes 5 ∼ 7 s. Furthermore, gradient-weighted class activation mapping (Grad-CAM) and uniform manifold approximation and projection (UMAP) explanatory algorithms were used to improve the interpretability and transparency of SqueezeNet. Thus, this study shows a new idea for FSA assisted with DL in meat freshness intelligent monitoring from development to deployment.

High sensitive Ratiometric fluorescent Aptasensor with AIE properties for Deoxynivalenol (DON) detection

An emerging fluorescent ratiometric aptasensor based on gold nanoclusters (AuNCs) with aggregation-induced emission (AIE) properties was prepared and studied for deoxynivalenol (DON) detection. The ratiometric aptasensor used red fluorescent AuNCs620 labelled with DON aptamer (Apt-AuNCs620) as an indicator and green fluorescent AuNCs519 modified by complementary DNA (cDNA) and magnetic beads (MBs) as internal reference, namely MBs-cDNA-AuNCs519. Under the optimal conditions, the aptasensor exhibited two good linear ranges of 0.1-50 and 50-5000 pg/mL for DON detection with coefficient of determination (R2) of 0.9937 and 0.9928, respectively, and the low detection limit (LOD) of 4.09 pg/mL was achieved. Furthermore, this aptasensor was feasible to detect DON in positive wheat samples, and the results were in line with those from HPLC and ELISA, thus providing a promising route to detect DON with high sensitivity in cereals, even for other mycotoxins by replacing the suitable aptamer and cDNA.

Elevating nanomaterial optical sensor arrays through the integration of advanced machine learning techniques for enhancing visual inspection of food quality and safety

Food quality and safety problems caused by inefficient control in the food chain have significant implications for human health, social stability, and economic progress and optical sensor arrays (OSAs) can effectively address these challenges. This review aims to summarize the recent applications of nanomaterials-based OSA for food quality and safety visual monitoring, including colourimetric sensor array (CSA) and fluorescent sensor array (FSA). First, the fundamental properties of various advanced nanomaterials, mainly including metal nanoparticles (MNPs) and nanoclusters (MNCs), quantum dots (QDs), upconversion nanoparticles (UCNPs), and others, were described. Besides, the diverse machine learning (ML) and deep learning (DL) methods of high-dimensional data obtained from the responses between different sensing elements and analytes were presented. Moreover, the recent and representative applications in pesticide residues, heavy metal ions, bacterial contamination, antioxidants, flavor matters, and food freshness detection were comprehensively summarized. Finally, the challenges and future perspectives for nanomaterials-based OSAs are discussed. It is believed that with the advancements in artificial intelligence (AI) techniques and integrated technology, nanomaterials-based OSAs are expected to be an intelligent, effective, and rapid tool for food quality assessment and safety control.

Developing a magnetic SERS nanosensor utilizing aminated Fe-Based MOF for ultrasensitive trace detection of organophosphorus pesticides in apple juice

The unreasonable use of organophosphorus pesticides leads to excessive pesticide residues in food, seriously threatening public health, and the potential of surface-enhanced Raman spectroscopy (SERS) technology, incorporating a metal-organic framework, is substantial for the rapid detection of trace pesticide residues. Here, a novel Fe3O4@NH2-MIL-101(Fe)@Ag (FNMA) SERS nanosensor was developed. Results indicated that the FNMA had a high enhancement factor of 1.53 × 108, a low limit of detection (LOD) of 4.55 × 10-12 M, and a relative standard deviation of 7.73 % for 4-nitrothiophenol, demonstrating its good SERS sensitivity and uniformity, and also possessed good storage stability for one month. In quantifying fenthion and methyl parathion in standard solutions and apple juice in the range of 0.05/0.02-20 mg/L, it showed LODs of 3.02 × 10-3 mg/L and 1.43 × 10-3 mg/L, and 0.0407 and 0.0075 mg/L, respectively, demonstrating potentials in ultrasensitive trace detection of pesticides in food.

A fluorescent Aptasensor based on magnetic-separation strategy with gold nanoclusters for Deoxynivalenol (DON) detection

A highly sensitive method based on MBs-cDNA@Apt-AuNCs519 was developed for deoxynivalenol (DON) detection in wheat. The MBs-cDNA@Apt-AuNCs519 was established using green emission gold nanoclusters (AuNCs519) with aggregation-induced emission properties as signal probes and combining amino-modified DON-aptamer (Apt), biotin-modified DNA strand (the partially complementary to Apt (cDNA)), and streptavidin-modified magnetic beads (MBs). The Apt-AuNCs519 were well connected with MBs-cDNA without DON but dissociated from MBs-cDNA@Apt-AuNCs519 with the addition of DON, leading to a noticeable reduction in the fluorescent intensity of the aptasensor. Moreover, this fluorescence aptasensor showed two linear relationships in the concentration range of 0.1-50 ng/mL and 50-5000 ng/mL with a limit of detection of 3.73 pg/mL with good stability, reproducibility and specificity. The results were consistent with high-performance liquid chromatography and enzyme-linked immunosorbent assay methods, further indicating the potential of this method for accurate trace detection of DON in wheat.

Nanotechnology improves the detection of bacteria: Recent advances and future perspectives

Nanotechnology has advanced significantly, particularly in biomedicine, showing promise for nanomaterial applications. Bacterial infections pose persistent public health challenges due to the lack of rapid pathogen detection methods, resulting in antibiotic overuse and bacterial resistance, threatening the human microbiome. Nanotechnology offers a solution through nanoparticle-based materials facilitating early bacterial detection and combating resistance. This study explores recent research on nanoparticle development for controlling microbial infections using various nanotechnology-driven detection methods. These approaches include Surface Plasmon Resonance (SPR) Sensors, Surface-Enhanced Raman Scattering (SERS) Sensors, Optoelectronic-based sensors, Bacteriophage-Based Sensors, and nanotechnology-based aptasensors. These technologies provide precise bacteria detection, enabling targeted treatment and infection prevention. Integrating nanoparticles into detection approaches holds promise for enhancing patient outcomes and mitigating harmful bacteria spread in healthcare settings.

Visible detection of chilled beef freshness using a paper-based colourimetric sensor array combining with deep learning algorithms

This study developed an innovative approach that combines a colourimetric sensor array (CSA) composed of twelve pH-response dyes with advanced algorithms, aiming to detect amine gases and assess the freshness of chilled beef. With the assistance of multivariate statistical analysis, the sensor array can effectively distinguish five amine gases and enable rapid quantification of trimethylamine vapour with a limit of detection (LOD) of 8.02 ppb and visually monitor the fresh levels of chilled beef. Moreover, the utilization of deep learning models (ResNet34, VGG16, and GoogleNet) for chilled beef freshness evaluation achieved an overall accuracy of 98.0 %. Furthermore, t-distributed stochastic neighbour embedding (t-SNE) visualized the feature extraction process and provided explanations to understand the classification process of deep learning. The results demonstrated that applying deep learning techniques in the process of pattern recognition of CSA can help in realizing the rapid, robust, and accurate assessment of chilled beef freshness.

Plasmonic cellulose microfilament assisted SERS detection in microfluidics

Limited by the narrow enhanced area of nanoscale on the metal surface, the sensitivity of surface-enhanced Raman spectroscopy (SERS) detection in solution is usually much lower than the detection in a solid substrate, which is dramatic in microfluidics for online detection. In this work, a cellulose microfilament embraced by Ag nanoparticles, called plasmonic cellulose microfilament, is located in a microchannel for SERS detection in microfluidics. Benefiting from the congestion caused by the plasmonic cellulose microfilament in a microchannel, the trace molecule in the solution is much easier to gather in Ag nanoparticles for Raman enhancement. To obtain high sensitivity, the structure of plasmonic cellulose microfilament is optimized. The SERS spectra collected in this novel microfluidics demonstrate that the plasmonic cellulose microfilament presents a high sensitivity at 10-13 M and good reproducibility in SERS detection. In addition, automatic identification of urea presence or absence was achieved based on deep learning (DL) here. The results show excellent diagnostic accuracy (99 %), which suggests that a fast, label-free urea screening tool can be developed. These results point out this SERS microfluidics with plasmonic cellulose microfilament has a great application prospective in online SERS detection with high sensitivity.

A durian-shaped multilayer core-shell SERS substrate for flow magnetic detection of pesticide residues on foods

A new type of durian-shaped Fe3O4@Au@Ag@Au (DFAAA) multilayer core-shell composite was prepared as an efficient surface-enhanced Raman scattering (SERS) substrate. The optimization process and SERS enhancement mechanism of the substrate were further explained with finite-difference time-domain simulation. The dense and uniform spiny array on the DFAAA surface had abundant "hot spots", greatly improving sensitivity, uniformity and reproducibility, with a Raman enhancement factor of 3.01 × 107 and storage-life of 30 d. A "flow magnetic detection method" was proposed to realize rapid and flexible detection of pesticide residues on the surface of different foods including fish and apple. The limit of detection of malachite green and thiram on the fish and apple surfaces were 0.13 and 0.18 ng/cm2, respectively. With its high SERS performance and good magnetic, the DFAAA possessed great application prospects as a facile SERS substrate for rapid and non-destructive detection of trace pesticide residues on foods.

Fluorescent polythymidine-templated copper nanoclusters aptasensor for sensitive detection of tropomyosin in processed shrimp products

Tropomyosin (TM) is the main allergen in shellfish. Developing a novel, simple and accurate method to track and detect TM in food products is necessary. In this work, a label-free fluorescent aptasensor based on polythymidine (poly(T))-templated copper nanoclusters (CuNCs) was designed for sensitive detection of TM in processed shrimp products. Magnetic beads (MBs), aptamer and cDNA were used to construct an MBs-aptamer@cDNA complex as a detection probe, and with the presence of TM, the poly(T)-templated CuNCs attached at the end of the cDNA as the fluorescent signal was released from the complex to turn on the fluorescence. Under optimal conditions, the poly(T)-templated CuNCs aptasensor achieved a linear range from 0.1 to 50 μg/mL (R2 = 0.9980), a low limit of detection of 0.0489 μg/mL and an excellent recovery percentage of 105.29%-108.91% in the complex food matrix, providing a new approach for food safety assurance.

A three-dimensional gold nanoparticles spherical liquid array for SERS sensitive detection of pesticide residues in apple

High-performance plasmonic substrates have recently attracted much research attention. Herein, a three-dimensional gold nanoparticles (AuNPs) spherical liquid array (SLA) with high "hot spots" and tunable nanometer gap by optimizing the proportion of AuNPs colloids over chloroform was synthesized based on a water-oil interfacial self-assembly strategy. The substrate demonstrated excellent surface-enhanced Raman scattering (SERS) performance using tetrathiafulvalene and rhodamine 6G (R6G) as probe molecules. With a simple extraction and soaking pretreatment process, the SLA exhibited high sensitivity for analysing triazophos on apple peels, with a limit of detection (LOD) of 0.005 µg/mL and recovery ranging from 96 to 110 %. Particularly, the chloroform produced an inherent characteristic peak at 665 cm-1, which was used as the internal standard to correct SERS signal fluctuation, leading to an improvement of the corresponding coefficient R2 from 0.97 to 0.99, thus improving the reproducibility. Therefore the SLA substrate possesses the potential for quantitative analysis of food contaminants.

Developments in food neonicotinoids detection: novel recognition strategies, advanced chemical sensing techniques, and recent applications

Neonicotinoid insecticides (NEOs) are a new class of neurotoxic pesticides primarily used for pest control on fruits and vegetables, cereals, and other crops after organophosphorus pesticides (OPPs), carbamate pesticides (CBPs), and pyrethroid pesticides. However, chronic abuse and illegal use have led to the contamination of food and water sources as well as damage to ecological and environmental systems. Long-term exposure to NEOs may pose potential risks to animals (especially bees) and even human health. Consequently, it is necessary to develop effective, robust, and rapid methods for NEOs detection. Specific recognition-based chemical sensing has been regarded as one of the most promising detection tools for NEOs due to their excellent selectivity, sensitivity, and robust interference resistance. In this review, we introduce the novel recognition strategies-enabled chemical sensing in food neonicotinoids detection in the past years (2017-2023). The properties and advantages of molecular imprinting recognition (MIR), host-guest recognition (HGR), electron-catalyzed recognition (ECR), immune recognition (IR), aptamer recognition (AR), and enzyme inhibition recognition (EIR) in the development of NEOs sensing platforms are discussed in detail. Recent applications of chemical sensing platforms in various food products, including fruits and vegetables, cereals, teas, honey, aquatic products, and others are highlighted. In addition, the future trends of applying chemical sensing with specific recognition strategies for NEOs analysis are discussed.

SERS immunoassay analysis of Escherichia coli and Staphylococcus aureus based on sandwich-structured complex probe and target-induced strand displacement

A direct and ultra-sensitive surface-enhanced Raman scattering (SERS) immunoassay method is introduced for the detection of Escherichia coli and Staphylococcus aureus. This methodology is based on a sandwich-structured complex probe (SCP) mechanism, combined with target-induced strand displacement. Moreover, by leveraging the amplified SERS signal from gold nanoparticles (AuNPs) corresponding to an increase in bacterial count, we achieve quantitative determination. The SCP demonstrates remarkable specificity, sensitivity, and anti-interference capability in bacterial detection. The detection limits for both bacterial strains are as low as 10 CFU/mL. In our selectivity tests, all peak intensities had standard deviations (n = 3) below 6%. Recoveries in normal human serum were 101-110% for E. coli and 96-101% for S. aureus. In milk, the recoveries were 102-105% for E. coli and 100-105% for S. aureus, respectively, demonstrating a high level of accuracy and resistance to interference. In addition, the SCP offers a dual-detection capability, enabling simultaneous diagnosis of multiple targets, which greatly simplifies the testing procedure. The findings underscore that this immunoassay platform fulfills the demand for rapid and precise pathogenic bacterial diagnosis, holding substantial potential for practical applications.

CRISPR/Cas12a and G-quadruplex DNAzyme-driven multimodal biosensor for visual detection of Aflatoxin B1

Aflatoxin B1 (AFB1) contamination severely threatens human and animal health, it is thus critical to construct a strategy for its rapid, accurate, and visual detection. Herein, a multimodal biosensor was proposed based on CRISPR/Cas12a cleaved G-quadruplex (G4) for AFB1 detection. Briefly, specific binding of AFB1 to the aptamer occupied the binding site of the complementary DNA (cDNA), and cDNA then activated Cas12a to cleave G4 into fragments. Meanwhile, the intact G4-DNAzyme could catalyze 3, 3', 5, 5'-tetramethylbenzidine (TMB) to form colourimetric/SERS/fluorescent signal-enhanced TMBox, and the yellow solution produced by TMBox under acidic conditions could be integrated with a smartphone application for visual detection. The colourimetric/SERS/fluorescent biosensor yielded detection limits of 0.85, 0.79, and 1.65 pg·mL-1, respectively, and was applied for detecting AFB1 in peanut, maize, and badam samples. The method is suitable for visual detection in naturally contaminated peanut samples and has prospective applications in the food industry.

Metal-Phenolic Networks for Chronic Wounds Therapy

Chronic wounds are recalcitrant complications of a variety of diseases, with pathologic features including bacterial infection, persistent inflammation, and proliferation of reactive oxygen species (ROS) levels in the wound microenvironment. Currently, the use of antimicrobial drugs, debridement, hyperbaric oxygen therapy, and other methods in clinical for chronic wound treatment is prone to problems such as bacterial resistance, wound expansion, and even exacerbation. In recent years, researchers have proposed many novel materials for the treatment of chronic wounds targeting the disease characteristics, among which metal-phenolic networks (MPNs) are supramolecular network structures that utilize multivalent metal ions and natural polyphenols complexed through ligand bonds. They have a flexible and versatile combination of structural forms and a variety of formations (nanoparticles, coatings, hydrogels, etc.) that can be constructed. Functionally, MPNs combine the chemocatalytic and bactericidal properties of metal ions as well as the anti-inflammatory and antioxidant properties of polyphenol compounds. Together with the excellent properties of rapid synthesis and negligible cytotoxicity, MPNs have attracted researchers' great attention in biomedical fields such as anti-tumor, anti-bacterial, and anti-inflammatory. This paper will focus on the composition of MPNs, the mechanisms of MPNs for the treatment of chronic wounds, and the application of MPNs in novel chronic wound therapies.

Research advances of SERS analysis method based on silent region molecules for food safety detection

Food safety is a critical issue that is closely related to people's health and safety. As a simple, rapid, and sensitive detection technique, surface-enhanced Raman scattering (SERS) technology has significant potential for food safety detection. Recently, researchers have shown a growing interest in utilizing silent region molecules for SERS analysis. These molecules exhibit significant Raman scattering peaks in the cellular Raman silent region between 1800 and 2800 cm-1 avoiding overlapping with the SERS spectrum of biological matrices in the range 600-1800 cm-1, which could effectively circumvent matrix effects and improve the SERS accuracy. In this review, the application of silent region molecules-based SERS analytical technique for food safety detection is introduced, detection strategies including label-free detection and labeled detection are discussed, and recent applications of SERS analysis technology based on molecules containing alkyne and nitrile groups, as well as Prussian blue (PB) in the detection of pesticides, mycotoxins, metal ions, and foodborne pathogens are highlighted. This review aims to draw the attention to the silent region molecules-based SERS analytical technique and to provide theoretical support for its further applications in food safety detection.

An interference-free SERS-based aptasensor for chlorpyrifos detection

In this study, we propose an interference-free SERS-based aptasensor for trace detection of chlorpyrifos (CPF) in real samples. In the aptasensor, gold nanoparticles coated with Prussian blue (Au@PB NPs) were employed as SERS tags to provide a sole and intense Raman emission at 2160 cm^-1, which could avoid overlapping with the Raman spectrum of the real samples in 600-1800 cm^-1 to improve the anti-matrix effect ability of the aptasensor. Under the optimum conditions, this aptasensor displayed a linear response for CPF detection in the range of 0.1-316 ng mL^-1 with a low detection limit of 0.066 ng mL^-1. In addition, the prepared aptasensor shows excellent application to determine CPF in cucumber, pear and river water samples. The recovery rates were highly correlated with high-performance liquid chromatography‒mass spectrometry (HPLC‒MS/MS). This aptasensor shows interference-free, specific and sensitive detection for CPF and offers an effective strategy for other pesticide residue detection.

Nanomaterial interfaces designed with different biorecognition elements for biosensing of key foodborne pathogens

Foodborne diseases caused by pathogen bacteria are a serious problem toward the safety of human life in a worldwide. Conventional methods for pathogen bacteria detection have several handicaps, including trained personnel requirement, low sensitivity, laborious enrichment steps, low selectivity, and long-term experiments. There is a need for precise and rapid identification and detection of foodborne pathogens. Biosensors are a remarkable alternative for the detection of foodborne bacteria compared to conventional methods. In recent years, there are different strategies for the designing of specific and sensitive biosensors. Researchers activated to develop enhanced biosensors with different transducer and recognition elements. Thus, the aim of this study was to provide a topical and detailed review on aptamer, nanofiber, and metal organic framework-based biosensors for the detection of food pathogens. First, the conventional methods, type of biosensors, common transducer, and recognition element were systematically explained. Then, novel signal amplification materials and nanomaterials were introduced. Last, current shortcomings were emphasized, and future alternatives were discussed.

Aggregates-based fluorescence sensing technology for food hazard detection: Principles, improvement strategies, and applications

Aggregates often exhibit modified or completely new properties compared with their molecular elements, making them an extraordinarily advantageous form of materials. The fluorescence signal change characteristics resulting from molecular aggregation endow aggregates with high sensitivity and broad applicability. In molecular aggregates, the photoluminescence properties at the molecular level can be annihilated or elevated, leading to aggregation-causing quenching (ACQ) or aggregation-induced emission (AIE) effects. This change in photoluminescence properties can be intelligently introduced in food hazard detection. Recognition units can combine with the aggregate-based sensor by joining the aggregation process, endowing the sensor with the high specificity of analytes (such as mycotoxins, pathogens, and complex organic molecules). In this review, aggregation mechanisms, structural characteristics of fluorescent materials (including ACQ/AIE-activated), and their applications in food hazard detection (with/without recognition units) are summarized. Because the design of aggregate-based sensors may be influenced by the properties of their components, the sensing mechanisms of different fluorescent materials were described separately. Details of fluorescent materials, including conventional organic dyes, carbon nanomaterials, quantum dots, polymers and polymer-based nanostructures and metal nanoclusters, and recognition units, such as aptamer, antibody, molecular imprinting, and host-guest recognition, are discussed. In addition, future trends of developing aggregate-based fluorescence sensing technology in monitoring food hazards are also proposed.

Multicolor aptasensors for pesticide multiresidues detection in agricultural products using bioorthogonal surface-enhanced Raman scattering tags

Realizing accurate pesticide multiresidue detection in a complex matrix is still a challenge for point-of-care sensing methods. Herein, we introduced background-free and multicolor aptasensors based on bioorthogonal surface-enhanced Raman scattering (SERS) tags and successfully applied them to analyze multiple pesticide residues. The excellent anti-interference and multiplex capability are due to the application of three bioorthogonal Raman reporters involving 4-ethenylbenzenamine (4-EBZM), Prussian blue (PB) and 2-amino-4-cyanopyridine (AMCP) with alkynyl and cyano groups, which demonstrated apparent Raman shift peaks at 1993 cm^-1, 2160 cm^-1, and 2264 cm^-1 in the biologically Raman-silent region, respectively. Ultimately, a detection range of 1-50 nM for acetamiprid, atrazine and malathion was achieved with detection limits of 0.39, 0.57 and 0.16 nM, respectively. The developed aptasensors were successfully used to determine pesticide residues in real samples. These proposed multicolor aptasensors offer an effective strategy for pesticide multiresidue detection with advantages of anti-interference, high specificity and high sensitivity.

Preparation of Fe3O4@UiO-66(Zr)@Ag NPs core-shell-satellite structured SERS substrate for trace detection of organophosphorus pesticides residues

Surface-enhanced Raman spectroscopy (SERS) technology has been revived and developed with the introduction of metal-organic frameworks (MOFs), while more valuable properties of MOFs for SERS substrates remain largely unexplored. This work constructed a new SERS substrate Fe₃O₄@UiO-66(Zr)@Ag nanoparticles (FUAs) with excellent SERS detection sensitivity, uniformity, reproducibility and stability, exhibiting a high Raman enhancement factor (5.62 × 10⁶), low limit of detection (LOD, 2.11 × 10^-11 M) and RSD (12.41 %) for 4-NBT, and maintaining 81 % SERS activity within 60 days. The FUAs took full advantage of the strong affinity of UiO-66(Zr) for organophosphorus pesticides (OPs) to realize trace OPs detection. The LODs of phoxim, triazophos and methyl parathion in apple juice were 0.041, 0.021 and 0.0031 mg/L, respectively, with good linearities ranging from 0.02 or 0.1-50 mg/L, meeting the requirements of the food control standards, indicating that the potentials and prospects of the FUAs SERS substrate for trace detecting OPs in foods.

Distinguishing pericarpium citri reticulatae of different origins using terahertz time-domain spectroscopy combined with convolutional neural networks

The geographical indication of pericarpium citri reticulatae (PCR) is very important in grading the quality and price of PCRs. Therefore, terahertz time-domain spectroscopy (THz-TDS) technology combined with convolutional neural networks (CNN) was proposed to distinguish PCRs of different origins without damage in this study. The one-dimensional CNN (1D-CNN) model with an accuracy of 82.99% based on spectral data processed with SNV was established. The two-dimensional image features were transformed from unprocessed spectral data using the gramian angular field (GAF), the Markov transition field (MTF) and the recurrence plot (RP), which were used to build a two-dimensional CNN (2D-CNN) model with an accuracy of 78.33%. Further, the CNN models with different fusion methods were developed for fusing spectra data and image data. In addition, the adding spectra and images based on the CNN (Add-CNN) model with an accuracy of 86.17% performed better. Eventually, the Add-CNN model based on ten frequencies extracted using permutation importance (PI) achieved the identification of PCRs from different origins. Overall, the current study would provide a new method for identifying PCRs of different origins, which was expected to be used for the traceability of PCRs products.

Recent Trends in SERS-Based Plasmonic Sensors for Disease Diagnostics, Biomolecules Detection, and Machine Learning Techniques

Surface-enhanced Raman spectroscopy/scattering (SERS) has evolved into a popular tool for applications in biology and medicine owing to its ease-of-use, non-destructive, and label-free approach. Advances in plasmonics and instrumentation have enabled the realization of SERS's full potential for the trace detection of biomolecules, disease diagnostics, and monitoring. We provide a brief review on the recent developments in the SERS technique for biosensing applications, with a particular focus on machine learning techniques used for the same. Initially, the article discusses the need for plasmonic sensors in biology and the advantage of SERS over existing techniques. In the later sections, the applications are organized as SERS-based biosensing for disease diagnosis focusing on cancer identification and respiratory diseases, including the recent SARS-CoV-2 detection. We then discuss progress in sensing microorganisms, such as bacteria, with a particular focus on plasmonic sensors for detecting biohazardous materials in view of homeland security. At the end of the article, we focus on machine learning techniques for the (a) identification, (b) classification, and (c) quantification in SERS for biology applications. The review covers the work from 2010 onwards, and the language is simplified to suit the needs of the interdisciplinary audience.

Discrimination of Pericarpium Citri Reticulatae in different years using Terahertz Time-Domain spectroscopy combined with convolutional neural network

Pericarpium Citri Reticulatae (PCR) in longer storage years possess higher medicinal values, but their differentiation is difficult due to similar morphological characteristics. Therefore, this study investigated the feasibility of using terahertz time-domain spectroscopy (THz-TDS) combined with a convolutional neural network (CNN) to identify PCR samples stored from 1 to 20 years. The absorption coefficient and refractive index spectra in the range of 0.2-1.5 THz were acquired. Partial least squares discriminant analysis, random forest, least squares support vector machines, and CNN were used to establish discriminant models, showing better performance of the CNN model than the others. In addition, the output data points of the CNN intermediate layer were visualized, illustrating gradual changes in these points from overlapping to clear separation. Overall, THz-TDS combined with CNN models could realize rapid identification of different year PCRs, thus providing an efficient alternative method for PCR quality inspection.

A SERS-Fluorescence dual-signal aptasensor for sensitive and robust determination of AFB1 in nut samples based on Apt-Cy5 and MNP@Ag-PEI

Food aflatoxin B1 (AFB1) contamination greatly threatens human health and its sensitive determination is imperative. In this study, a surface-enhanced Raman scattering (SERS) and fluorescence dual-signal aptasensor was constructed for sensitive AFB1 detection in peanuts, walnuts, and almonds samples. Fluorescent dye cy5 was used as fluorophore and Raman reporter, while polyethyleneimine modified Ag coating magnetic nanoparticles (MNP@Ag-PEI) were utilized to absorb the cy5 modified aptamer (apt-cy5). Results indicated that linear ranges of 0.001-1000 ng/mL and 0.2-20,000 ng/mL with detection limits of 0.45 pg/mL and 0.135 ng/mL for the SERS and fluorescence methods were obtained, respectively, and AFB1 detection in the nut samples using the aptasensor achieved satisfactory recoveries of 95.2%-108.6% for SERS and 94.7%-109.7% for fluorescence. Compared with other mono signal detection, the established aptasensor facilely fused the merits of the two signals and improved the detection accuracy and flexibility.

Analysis and detection using novel terahertz spectroscopy technique in dietary carbohydrate-related research: Principles and application advances

As one of the main functional substances, carbohydrates account for a large proportion of the human diet. Conventional analysis and detection methods of dietary carbohydrates and related products are destructive, time-consuming, and labor-intensive. In order to improve the efficiency of measurement and ensure food nutrition and consumer health, rapid and nondestructive quality evaluation techniques are needed. In recent years, terahertz (THz) spectroscopy, as a novel detection technology with dual characteristics of microwave and infrared, has shown great potential in dietary carbohydrate analysis. The current review aims to provide an up-to-date overview of research advances in using the THz spectroscopy technique in analysis and detection applications related to dietary carbohydrates. In the review, the principles of the THz spectroscopy technique are introduced. Advances in THz spectroscopy for quantitative and qualitative analysis and detection in dietary carbohydrate-related research studies from 2013 to 2022 are discussed, which include analysis of carbohydrate concentrations in liquid and powdery foods, detection of foreign body and chemical residues in carbohydrate food products, authentication of natural carbohydrate produce, monitoring of the fermentation process in carbohydrate food production and examination of crystallinity in carbohydrate polymers. In addition, applications in dietary carbohydrate-related detection research using other spectroscopic techniques are also briefed for comparison, and future development trends of THz spectroscopy in this field are finally highlighted.

Mesoporous silica coated core-shell nanoparticles substrate for size-selective SERS detection of chloramphenicol

With the growing popularity of the non-destructive technique, surface-enhanced Raman spectroscopy (SERS) demands a highly sensitive and reproducible plasmonic nanoparticles substrate. In this study, a novel bimetallic core-shell nanoparticles (Au@Ag@mSiO₂NP) substrate consisting of a gold core, silver shell, and a mesoporous silica coating was synthesized. The mesoporous coating structure was created by employing template molecules such as surfactant and their subsequent removal allowing selective screening based on the size of analyte molecules. Results showed that the plasmonic substrate could selectively enhance small molecules by preventing large macromolecules to reach the exciting zone of the substrate core, achieving the detection of chloramphenicol in milk samples with a detection limit of 6.68 × 10^-8 M. Moreover, the mesoporous coating provided additional stability to the Au@Ag nanoparticles, leading to the reusability of the substrate. Thus, this work offered a simple and smart Au@Ag@mSiO₂NP substrate for effective SERS detection of analytes.

Ag Nanoparticles@Agar Gel as a 3D Flexible and Stable SERS Substrate with Ultrahigh Sensitivity

Flexible surface-enhanced Raman scattering (SERS) substrates have become one of the research hot spots due to the facile sampling by swabbing or wrapping on rough surfaces and the sensitive and nondestructive detection of contaminants. In this work, we proposed a simple and fast in situ reduction method to prepare Ag nanoparticles (Ag NPs) composited agar hydrogel (Ag NPs@Agar) flexible SERS substrate. Owing to the three-dimensional (3D) structure, good hydrophilicity and adsorption of the agar hydrogel, Ag NPs were grown uniformly in the 3D cross-linked structure. The distribution density of Ag NPs was further increased by the volume shrinkage when the hydrogel was dried in air. This high density and uniformly distribution of Ag NPs produced a large number of highly active SERS regions. In addition, the sensitivity of Ag NPs@Agar was further improved with the assistance of hydrophilic agar gel, which can trap the probe molecules into highly active SERS areas. The SERS results showed that the substrate can be used to detect dye molecules (rhodamine 6G), the minimum detectable concentration was 10^-15 M, the relative standard deviation tested at 18 different positions was only 7.58%, and the intensity of the characteristic peak at 611 cm^-1 decreased only about 10% after 49 days of storage, demonstrating the superior stability. Moreover, the Ag NPs@Agar substrate also could successfully achieve the micro-trace detection of melamine and sodium penicillin G in Xinjiang specialty camel milk powder. The above available results show that the prepared flexible Ag NPs@Agar SERS substrates possess potentials for the illegal additives and antibiotics in food safety analysis.

Surface-enhanced Raman spectroscopy combined with stable isotope probing to assess the metabolic activity of Escherichia coli cells in chicken carcass wash water

Rapid evaluation of the metabolic activity of microorganisms is crucial in the assessment of the disinfection ability of various antimicrobial agents in the food industry. In this study, surface-enhanced Raman spectroscopy combined with isotope probing was employed for the analysis of the disinfection of single bacterial cells in the chicken carcass wash water. The Raman signals from single Escherichia coli O157:H7 cells were enhanced by in situ synthesis of silver nanoparticles. The ΔCD of the cells grown in presence of 0.5% hydrogen peroxide and 50 ppm chlorine was 5.86 ± 1.86% and 5.1 ± 2.3%, respectively, which showed significant reduction compared with cells grown in the absence of disinfecting agents (19.86 ± 2.51%) after 2 h of incubation. The study proved that the proposed method had the potential to assess the metabolic activity of microorganisms in other food products and optimize the disinfection process.

Visualization and quantification of content and hydrogen bonding state of water in apple and potato cells by confocal Raman microscopy: A comparison study

Water is the most abundant component in fresh fruit and vegetables and its distribution and hydrogen bonding state in cells has a significant influence on food processing. In the current study, an improved method based on our earlier studies was developed to directly visualize the spatial distribution of content and hydrogen bonding state of water in apple and potato cells for the first time and the difference in water distribution in these cells was compared. Additionally, based on the distribution images of content and hydrogen bonding state of water in different regions in apple and potato tissues, the total water and free water contents, and the hydrogen bonding state of free water were quantified and compared with those obtained by nuclear magnetic resonance and Marinchik methods, demonstrating that the method could be successfully used for quantifying the content and hydrogen bonding state of water in fruit and vegetable cells.

Analyzing macromolecular composition of E. Coli O157:H7 using Raman-stable isotope probing

Metabolic dynamics of bacterial cells is needed for understanding the correlation between changes in environmental conditions and cell metabolic activity. In this study, Raman spectroscopy combined with deuterium labelling was used to analyze the metabolic activity of a single Escherichia coli O157:H7 cell. The incorporation of deuterium from heavy water into cellular biomolecules resulted in the formation of carbon-deuterium (CD) peaks in the Raman spectra, indicating the cell metabolic activity. The broad vibrational peaks corresponding to CD and CH peaks encompassed different specific shifts of macromolecules such as protein, lipids, and nucleic acid. The utilization of tryptophan and oleic acid by the cell as the sole carbon source led to changes in cell lipid composition, as indicated by new peaks in the second derivative spectra. Thus, the proposed method could semi-quantitatively determine total metabolic activity, macromolecule specific identification, and lipid and protein metabolism in a single cell.

Specific and quantitative detection of bacteria based on surface cell imprinted SERS mapping platform

Non-specificity and poor quantitative ability are the main challenges in surface-enhanced Raman scattering (SERS) technique, especially for the detection of bacteria in real samples. In this study, we presented a surface cell imprinted SERS mapping platform which is competent for the specific and quantitative detection of bacteria. The platform based on the fabrication of a surface cell imprinted substrate (SCIS) by which Escherichia coli (E. coli) can be captured and labelled by SERS tags which produces strong characteristic signal to indicate the capture of targets. We highlighted the specificity of this platform in the detection of E. coli, by comparing the performances toward Salmonella paratyphoid A, Bacillus subtilis, Enterococcus faecalis and Staphylococcus aureus. Upon integrating with SERS mapping technique, the platform displayed good quantitative ability toward E. coli with a wide linear range from 10² to 10⁸ CFU/mL and a low detection limit of ∼1.35 CFU/mL. Moreover, this novel SERS analysis platform was proved to be effective for E. coli detection in real probiotic beverage and chicken breast meat samples. By fabricating different SCISs, this platform can be replicated for the detection of other bacteria, which provides a promising application for real sample testing.

2D nanomaterial sensing array using machine learning for differential profiling of pathogenic microbial taxonomic identification

An integrated custom cross-response sensing array has been developed combining the algorithm module’s visible machine learning approach for rapid and accurate pathogenic microbial taxonomic identification. The diversified cross-response sensing array consists of two-dimensional nanomaterial (2D-n) with fluorescently labeled single-stranded DNA (ssDNA) as sensing elements to extract a set of differential response profiles for each pathogenic microorganism. By altering the 2D-n and different ssDNA with different sequences, we can form multiple sensing elements. While interacting with microorganisms, the competition between ssDNA and 2D-n leads to the release of ssDNA from 2D-n. The signals are generated from binding force driven by the exfoliation of either ssDNA or 2D-n from the microorganisms. Thus, the signal is distinguished from different ssDNA and 2D-n combinations, differentiating the extracted information and visualizing the recognition process. Fluorescent signals collected from each sensing element at the wavelength around 520 nm are applied to generate a fingerprint. As a proof of concept, we demonstrate that a six-sensing array enables rapid and accurate pathogenic microbial taxonomic identification, including the drug-resistant microorganisms, under a data size of n = 288. We precisely identify microbial with an overall accuracy of 97.9%, which overcomes the big data dependence for identifying recurrent patterns in conventional methods. For each microorganism, the detection concentration is 10⁵ ~ 10⁸ CFU/mL for Escherichia coli, 10² ~ 10⁷ CFU/mL for E. coli-β, 10³ ~ 10⁸ CFU/mL for Staphylococcus aureus, 10³ ~ 10⁷ CFU/mL for MRSA, 10² ~ 10⁸ CFU/mL for Pseudomonas aeruginosa, 10³ ~ 10⁸ CFU/mL for Enterococcus faecalis, 10² ~ 10⁸ CFU/mL for Klebsiella pneumoniae, and 10³ ~ 10⁸ CFU/mL for Candida albicans. Combining the visible machine learning approach, this sensing array provides strategies for precision pathogenic microbial taxonomic identification.

Detection of Bioactive Metabolites in Escherichia Coli Cultures Using Surface-Enhanced Raman Spectroscopy

Detection of bioactive metabolites produced by bacteria is important for identifying biomarkers for infectious diseases. In this study, a surface-enhanced Raman spectroscopy (SERS)-based technique was developed for the detection of bioactive metabolite indole produced by Escherichia coli (E. coli) in biological media. The use of highly sensitive Au@Ag core-shell nanoparticles resulted in the detection of indole concentration as low as 0.0886 mM in standard solution. The supplementation of growth media with 5 mM of exogenous tryptophan resulted in the production of a maximum yield of indole of 3.139 mM by E. coli O157:H7 at 37 °C. The growth of bacterial cells was reduced from 47.73 × 10⁸ to 1.033 × 10⁶ CFU/mL when the cells were grown in 0 and 10 mM exogenous tryptophan, respectively. The amount of indole in the Luria-Bertani (LB) media had an inverse correlation with the growth of cells, which resulted in a three-log reduction in the colony-forming unit when the indole concentration in the media was 20 times higher than normal. This work demonstrates that SERS is an effective and highly sensitive method for rapid detection of bioactive metabolites in biological matrix.

Recent Advances in Aptasensors For Rapid and Sensitive Detection of Staphylococcus Aureus

The infection of Staphylococcus aureus (S.aureus) and the spread of drug-resistant bacteria pose a serious threat to global public health. Therefore, timely, rapid and accurate detection of S. aureus is of great significance for food safety, environmental monitoring, clinical diagnosis and treatment, and prevention of drug-resistant bacteria dissemination. Traditional S. aureus detection methods such as culture identification, ELISA, PCR, MALDI-TOF-MS and sequencing, etc., have good sensitivity and specificity, but they are complex to operate, requiring professionals and expensive and complex machines. Therefore, it is still challenging to develop a fast, simple, low-cost, specific and sensitive S. aureus detection method. Recent studies have demonstrated that fast, specific, low-cost, low sample volume, automated, and portable aptasensors have been widely used for S. aureus detection and have been proposed as the most attractive alternatives to their traditional detection methods. In this review, recent advances of aptasensors based on different transducer (optical and electrochemical) for S. aureus detection have been discussed in details. Furthermore, the applications of aptasensors in point-of-care testing (POCT) have also been discussed. More and more aptasensors are combined with nanomaterials as efficient transducers and amplifiers, which appears to be the development trend in aptasensors. Finally, some significant challenges for the development and application of aptasensors are outlined.

Photosensitized Peroxidase Mimicry at the Hierarchical 0D/2D Heterojunction-Like Quasi Metal-Organic Framework Interface for Boosting Biocatalytic Disinfection

Metal-organic frameworks (MOFs) are a versatile toolbox for the bioinspired design of nanozymes for antibacterial applications and beyond, however, designing a nanozyme by the hierarchical quasi-MOF scheme remains largely unpracticed. This work exemplifies the preferential structure-activity correlation of a bimetallic quasi-MOF (Q-MOF_Ce0.5 ) among three series of MOF-derived peroxidase (POD) mimics. The biomimetic quasi-MOF_Ce0.5 nanosheets accommodate both oxygen vacancy-coupled multivalent redox cycles and photosensitive energy band layout, benefiting from the hierarchical heterojunction-like 0D/2D interface featuring isolated nodes-derived CeOCu sites upon the 2D decarboxylated MOF scaffold. These integrated unique merits enable the POD-like Q-MOF_Ce0.5 to generate sustained reactive oxygen species to effectively eradicate the surface-adhered bacteria under visible light, resulting in significant inactivation of Escherichia coli (99.74 %) and Staphylococcus aureus (99.35%) in vitro, and potent disinfection of skin wounds in vivo in safe and on-demand manners. It is hoped that this work can intensify the interventions of MOF nanozymes against the microbial world.

Novel graphene oxide/polymer composite membranes for the food industry: structures, mechanisms and recent applications

The membrane can not only be used as food packaging, but also for the separation, fractionation and recovery of food ingredients. Graphene oxide (GO) sheets are a two-dimensional (2 D) material with a unique structure that exhibit excellent mechanical properties, biocompatibility, and flexibility. The corporation of polymer matrix membrane with GO can significantly improve the permeability, selectivity, and antibacterial activity. In this review, the chemical structures of GO, GO membranes and GO/polymer composite membranes are introduced, the permeation mechanisms of molecules through the membranes are discussed and key factors affecting the permeability are presented in detail. In addition, recent applications in the food industry for filtration, bioreactions and active food packaging are analyzed, and limitations and future trends of GO membranes development are also highlighted. GO/polymer composite membranes exhibit excellent permeability, selectivity and strong barrier properties against bacterial and gas permeation. However, current food material filtration and packaging applications of GO/polymer composite membranes are still in the laboratory stage. Future work can focus on the development of large scale uniformly sized GO production, the homogeneous distribution and tight combination of GO in polymer matrixes, the sensing function of GO in packaging, and the verification method of GO toxicology.

Interfacing metal-polyphenolic networks upon photothermal gold nanorods for triplex-evolved biocompatible bactericidal activity

Gold nanorods (GNRs) outstand in photothermal disinfection but are faced with severe surface chemistry and dose relevant biotoxicity. Herein, a naturally green building block, metal-phenolic networks (MPNs), was employed to functionalize GNRs via coordination reaction, yielding a tunable and biocompatible core-shell photothermal nano-bactericide (GNRs@MPNs). The bioactive GNRs@MPNs built with iron and polyphenols (tannic acid, epigallocatechin gallate, and procyanidins) exhibited superior light-to-heat conversion efficiencies with η = 29.29-44.00%, remarkably preceding that of GNRs (η = 12.24%), which could rapidly ablate 99.8% of Escherichia coli O157: H7 and 98.6% of Staphylococcus aureus bacteria in relatively low efficacy doses (10 ppm of Au). Moreover, local heat triggered by GNRs@MPNs accelerated the healing of the cutaneous wound of a mice model infected by methicillin-resistant S. aureus. The facile synthesis, photothermal synergy, polyphenolic bioactivity, and significantly low efficacy dose of GNRs@MPNs empower them satisfactory efficiency and biosafety in the future broad-spectrum photothermal sterilization applications.

Ti3C2Tx MXenes loaded with Au nanoparticle dimers as a surface-enhanced Raman scattering aptasensor for AFB1 detection

Mycotoxin B1 (AFB1) contamination in agricultural products pose a deadlydangertoanimal and human health and its rapid and reliable detection is thus very important. Herein, a ratiometric surface-enhanced Raman scattering (SERS) aptasensor for AFB1 detection was developed, in which 1,2-bis(4-pyridyl) ethylene (BPE) was used to trigger the assembly of Au nanoparticle dimers (AuNP dimers) and form intensive SERS "hot spots", and MXenes nanosheets could load aptamer-modified AuNP dimers due to the hydrogen bonding and the chelation between the phosphate groups of aptamers and the Ti ion of MXenes. With the presence of AFB1 preferentially binding to AFB1 aptamer, AuNP dimers were separated from MXenes nanosheets, leading to a decrease in SERS intensity. Regression analysis in the range from 0.001 to 100 ng·mL^-1 showed the limit of detection (LOD) being 0.6 pg·mL^-1 in standard solution, indicating that the great prospects of the AuNP dimers/MXenes SERS substrate for detecting AFB1.

Highly sensitive multiplex detection of foodborne pathogens using a SERS immunosensor combined with novel covalent organic frameworks based biologic interference-free Raman tags

Rapid and reliable multiplex detection of foodborne pathogens is in great demand for ensuring food safety and preventing foodborne diseases. In this study, we developed a highly sensitive SERS immunosensor for the simultaneous detection of multiple foodborne pathogens. Novel biologic interference-free Raman tags synthesized by using the covalent organic frameworks (COF) TBDP as nanocontainer to load biologic interference-free Raman reporters and specific antibodies for interested targets were used to convert and amplify signals of foodborne pathogens. In addition, lectin functionalized magnetic nanoparticles (MNPs@Con A) which could efficiently bind to the carbohydrate constituents on the surface of pathogens were prepared to capture and isolate multiple pathogens simultaneously. The recognition of the target foodborne pathogen impels the generation of sandwich-like composites of MNPs@Con A/pathogen/TBDP@Raman tags, and these composites could be quickly separated from the sample matrix with the assistance of an external magnet. Besides, a mass of Raman reporters was released by eluting the collected MNPs@Con A/pathogen/TBDP@Raman tags composites. Combined with a portable Raman system, characteristic Raman signals (2271 and 2113 cm^-1) of the occupied reporters located at the biologic interference-free region were observed and used for the simultaneous detection of two different foodborne pathogenic strains. And an equal limit of detection of 10¹ CFU/mL was achieved for each strain. This strategy provides new insight into the application of SERS in the detection of pathogenic bacteria.

A dual AE-GAN guided THz spectral dehulling model for mapping energy and moisture distribution on sunflower seed kernels

Energy and moisture contents are important food chemical attributes. In the current study, a nondestructive Terahertz (THz) time-domain imaging system was first time used for evaluating the energy and moisture distributions of sunflower seed kernels inside shells. For this task, a dual autoencoders (AE)-generative adversarial nets (GAN) spectral dehulling semi-supervised model was developed. The model could automatically learn the kernel information from the latent representations of the spectra of the intact seeds through adversarial learning to achieve feature disentanglement. Results indicated that the generated kernel images had similar features to the original kernel images and high-quality chemical distribution maps for energy and moisture contents of sunflower seed kernels inside shells were successfully obtained. As the current method took the advantage of the characteristics of THz imaging and selected a suitable deep learning algorithm, it has the potential to generalize for imaging other chemical substances of other dry shelled seeds or biological samples (moisture content and thickness below 15% and 5 mm, respectively).