Terahertz fingerprint characterization of 2,4-dichlorophenoxyacetic acid and its enhanced detection in food matrices combined with spectral baseline correction

Distinguishing different proteins based on terahertz spectra by visual geometry group 16 neural network

Detecting different kinds of proteins is of great significance for medical diagnosis, biological research, and other fields. We combine both terahertz (THz) absorption and refractive index spectra with the visual geometry group 16 (VGG-16) neural network to intelligently identify four proteins, namely albumin, collagen, pepsin, and pancreatin in this study. The THz absorption-refractive index spectra of the proteins were converted to two-dimensional image features by the Grassia angular summation field (GASF) method and used as a dataset, which enabled the VGG-16 model to achieve 98.8% accuracy in distinguishing the four proteins. We also compared the VGG-16 model with other machine learning models, which demonstrate that it has better performance. Overall, the VGG-16 neural network transfer learning technique proposed in this study can quickly and accurately achieve the identification of different kinds of proteins. This research might have potentially important applications in biotechnology fields, such as biosensors, biopharmaceuticals, and medicine.

Resonance features integration of multiple terahertz metamaterials sensors for qualification and quantification of trace fluoroquinolone antibiotics

BACKGROUND

The residual fluoroquinolone antibiotics (FQs) in the environment and food has raised public concerns over their potential impact on human health. Terahertz metamaterial sensors (TMSs) have garnered significant attention due to their capability to enhance the interaction between terahertz waves and antibiotic molecules, enabling the detection of trace antibiotics. However, conventional quantitative and qualitative methods based on TMSs suffer from low accuracy and cumbersome processes, respectively. Herein, this work proposed a novel approach that reconstructed optimal terahertz response features of different TMSs with machine learning algorithms, which allowed for analysis of three similar trace FQs with enhanced accuracy.

RESULTS

The prepared three patterned TMSs exhibited different resonance responses, which varied with changes in FQs types and concentrations. The resonance peak features of the three TMSs were fused to construct the resonance peak feature matrix (W0) and combined with the K-Nearest Neighbor (KNN) algorithm to build the W0-KNN classification model. The interval feature matrix was constructed by optimizing and expanding the resonance peak feature width. The optimal resonance peak interval feature matrix (Wt) was combined with Gaussian process regression (GPR) algorithms with different kernel functions to build the Wt-GPR quantitative model. The results showed that W0-KNN achieved 100 % classification accuracy for the three FQs. Wt-GPR exhibited high quantitative accuracy for all three FQs with the determination coefficient (R2) of 0.94-0.98, and root mean square error (RMSE) of 6.4085-10.6540. The results of Wt-GPR with different kernel functions had small fluctuations, demonstrating high stability in predictive performance.

SIGNIFICANCE

Reconstructing features from multi-TMSs in combination with machine learning algorithms enables rapid, precise, and reliable qualitative and quantitative analysis of trace FQs. Our research introduces innovative concepts and methodologies to detect trace FQs using TMS-based sensors, paving the way for future applications of TMS in the biomolecular sensing and detection.

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.

Comprehensive Similarity Algorithm and Molecular Dynamics Simulation-Assisted Terahertz Spectroscopy for Intelligent Matching Identification of Quorum Signal Molecules (N-Acyl-Homoserine Lactones)

To investigate the mechanism of aquatic pathogens in quorum sensing (QS) and decode the signal transmission of aquatic Gram-negative pathogens, this paper proposes a novel method for the intelligent matching identification of eight quorum signaling molecules (N-acyl-homoserine lactones, AHLs) with similar molecular structures, using terahertz (THz) spectroscopy combined with molecular dynamics simulation and spectral similarity calculation. The THz fingerprint absorption spectral peaks of the eight AHLs were identified, attributed, and resolved using the density functional theory (DFT) for molecular dynamics simulation. To reduce the computational complexity of matching recognition, spectra with high peak matching values with the target were preliminarily selected, based on the peak position features of AHL samples. A comprehensive similarity calculation (CSC) method using a weighted improved Jaccard similarity algorithm (IJS) and discrete Fréchet distance algorithm (DFD) is proposed to calculate the similarity between the selected spectra and the targets, as well as to return the matching result with the highest accuracy. The results show that all AHL molecular types can be correctly identified, and the average quantization accuracy of CSC is 98.48%. This study provides a theoretical and data-supported foundation for the identification of AHLs, based on THz spectroscopy, and offers a new method for the high-throughput and automatic identification of AHLs.

Predicting wheat gluten concentrations in potato starch using GPR and SVM models built by terahertz time-domain spectroscopy

The purpose of this study was for the first time to explore the feasibility of terahertz (THz) spectral imaging for the detection of gluten contents in food samples. Based on the obtained 80 THz spectrum data, Gaussian process regression (GPR) and support vector machine (SVM) models were established to predict wheat gluten concentrations in 40 potato starch mixture samples. The prediction performances of GPR and SVM obtained were R2 = 0.859 and RMSE = 0.070, and R2 = 0.715 and RMSE = 0.101 in the gluten concentration range of 1.3%-100%, respectively, showing that the linear SVM algorithm had better prediction performance. The results indicated that THz spectral imaging combined with GPR could be used to predict the gluten content in food samples. It is thus hoped that this research should provide a novel technique for gluten content detection to ensure gluten-free food samples.

Terahertz determination of imidacloprid in soil based on a metasurface sensor

Pesticides in soil are continuously one of the most studied analytes due to their environmental and human health effects. Thus the detection of pesticides in soil is an important means to control and assess soil quality. Here, we theoretically and experimentally present a novel method for the determination of imidacloprid in soil by using a metasurface sensor operating at terahertz frequencies. The metasurface shows a resonance peak at 880 GHz and the electric field at the peak is strongly localized and concentrated in the gap of split I-shaped resonator. The detection of complex refractive index shows that the position and the transmittance of resonance peak are depend on the change in the complex refractive index. The measurement of imidacloprid concentration in soil demonstrates that both the frequency shift and the transmittance change at peak increase almost linearly with the increasing of imidacloprid concentration ranging from 0.25% to 2%. In this case, the frequency shift reaches 97 GHz and the transmittance change at peak is as high as 30.9%. Our work enables the determination of imidacloprid in soil at terahertz frequencies with good reliability and high sensitivity, showing the potential application of terahertz spectroscopy in environmental monitoring.

A Review of Recent Advances for the Detection of Biological, Chemical, and Physical Hazards in Foodstuffs Using Spectral Imaging Techniques

Traditional methods for detecting foodstuff hazards are time-consuming, inefficient, and destructive. Spectral imaging techniques have been proven to overcome these disadvantages in detecting foodstuff hazards. Compared with traditional methods, spectral imaging could also increase the throughput and frequency of detection. This study reviewed the techniques used to detect biological, chemical, and physical hazards in foodstuffs including ultraviolet, visible and near-infrared (UV-Vis-NIR) spectroscopy, terahertz (THz) spectroscopy, hyperspectral imaging, and Raman spectroscopy. The advantages and disadvantages of these techniques were discussed and compared. The latest studies regarding machine learning algorithms for detecting foodstuff hazards were also summarized. It can be found that spectral imaging techniques are useful in the detection of foodstuff hazards. Thus, this review provides updated information regarding the spectral imaging techniques that can be used by food industries and as a foundation for further studies.

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.

High-Precision Automatic Identification of Fentanyl-Related Drugs by Terahertz Spectroscopy with Molecular Dynamics Simulation and Spectral Similarity Mapping

Fentanyl is a potent opioid analgesic with high bioavailability. It is the leading cause of drug addiction and overdose death. To better control the abuse of fentanyl and its derivatives, it is crucial to develop rapid and sensitive detection methods. However, fentanyl-related substrates undergo similar molecular structures resulting in similar properties, which are difficult to be identified by conventional spectroscopic methods. In this work, a method for the automatic identification of 8 fentanyl-related substances with similar spectral characteristics was developed using terahertz (THz) spectroscopy coupled with density functional theory (DFT) and spectral similarity mapping (SSM). To characterize the THz fingerprints of these fentanyl-related samples more accurately, the method of baseline estimation and denoising with sparsity was performed before revealing the unique molecular dynamics of each substance by DFT. The SSM method was proposed to identify these fentanyl analogs based on weighted spectral cosine–cross similarity and fingerprint discrete Fréchet distance, generating a matching list by stepwise searching the entire spectral database. The top matched list returned the identification results of the target fentanyl analogs with accuracies of 94.48~99.33%. Results from this work provide algorithms’ increased reliability, which serves as an artificial intelligence-based tool for high-precision fentanyl analysis in real-world samples.

A Novel Method for Carbendazim High-Sensitivity Detection Based on the Combination of Metamaterial Sensor and Machine Learning

Benzimidazole fungicide residue in food products poses a risk to consumer health. Due to its localized electric-field enhancement and high-quality factor value, the metamaterial sensor is appropriate for applications regarding food safety detection. However, the previous detection method based on the metamaterial sensor only considered the resonance dip shift. It neglected other information contained in the spectrum. In this study, we proposed a method for highly sensitive detection of benzimidazole fungicide using a combination of a metamaterial sensor and mean shift machine learning method. The unit cell of the metamaterial sensor contained a cut wire and two split-ring resonances. Mean shift, an unsupervised machine learning method, was employed to analyze the THz spectrum. The experiment results show that our proposed method could detect carbendazim concentrations as low as 0.5 mg/L. The detection sensitivity was enhanced 200 times compared to that achieved using the metamaterial sensor only. Our present work demonstrates a potential application of combining a metamaterial sensor and mean shift in benzimidazole fungicide residue detection.

Paper-based multicolor sensor for on-site quantitative detection of 2,4-dichlorophenoxyacetic acid based on alkaline phosphatase-mediated gold nanobipyramids growth and colorimeter-assisted method for quantifying color

On-site quantitative analysis of pesticides is important for food safety. Colorimetric gold nanobipyramids (AuNBPs) sensors are powerful methods for on-site detection. However, a single quantitative method and the instability of AuNBPs in solution limit the practicability of those sensors. Here, a paper-based multicolor AuNBPs sensor involved a colorimeter-assisted method for quantifying color was developed for quantitative detection of 2,4-dichlorophenoxyacetic acid (2,4-D), a common herbicide. The novelty of this study lies in developing a general paper-based quantitative on-site method (PQOM) for colorimetric AuNBPs sensors. Firstly, a paper-based analytical device (PAD) consisting of a nylon membrane, absorbent cotton layers, and two acrylic plates was fabricated to deposit AuNBPs. We demonstrated the PAD could improve the stability of AuNBPs and the detection sensitivity of AuNBPs sensors. Then, a handheld colorimeter was first used to quantify the color change of AuNBPs on the PAD based on the CIELab color space. Finally, as proof of concept, the PQOM was successfully employed to quantify 2,4-D by combining with an alkaline phosphatase-mediated AuNBPs growth method. In this method, 2,4-D specifically inhibited alkaline phosphatase activity to suppress the generation of l-ascorbic acid, thereby mediating AuNBPs growth. The developed sensor exhibited seven 2,4-D concentration-related colors and detected as low as 50 ng mL-1 2,4-D by naked-eye observation and 18 ng mL-1 2,4-D by a colorimeter. It was applied to detect 2,4-D in the spiked rice and apple samples with good recovery rates (91.8-112.0%) and a relative standard deviation (n = 5) < 5%. The success of this study provides a sensing platform for quantifying 2,4-D on site.

Simultaneous quantitative determination of low-concentration ternary pesticide mixtures in wheat flour based on terahertz spectroscopy and BPNN

Terahertz spectroscopy has been widely applied in the quantitative analysis of pesticides, however, it still encounters challenge pursuing high prediction accuracy in multi-component mixtures analysis with ultra-low concentration. Here, back propagation neural network (BPNN) was applied on the determination of ternary pesticide mixtures in wheat flour. By spectral pre-processing and model parameter optimization, high-quality spectra and complete network frame was achieved. On this basis, a novel wavelength selection method was presented and the most efficient peak width was given. Our method here achieved the optimal results, the correlation coefficient of prediction sets (R_P) were 0.9913, 0.9948, 0.9923, and corresponding root mean square error (RMSE) were 0.0211%, 0.0176%, 0.0191%. More importantly, the concentration of pesticides in this study was extremely low compared with similar quantitative analysis based on terahertz spectroscopy, which can promote the application of this technology into actual production.

Pesticide detection combining the Wasserstein generative adversarial network and the residual neural network based on terahertz spectroscopy

Feature extraction is a key factor to detect pesticides using terahertz spectroscopy. Compared to traditional methods, deep learning is able to obtain better insights into complex data features at high levels of abstraction. However, reports about the application of deep learning in THz spectroscopy are rare. The main limitation of deep learning to analyse terahertz spectroscopy is insufficient learning samples. In this study, we proposed a WGAN-ResNet method, which combines two deep learning networks, the Wasserstein generative adversarial network (WGAN) and the residual neural network (ResNet), to detect carbendazim based on terahertz spectroscopy. The Wasserstein generative adversarial network and pretraining model technology were employed to solve the problem of insufficient learning samples for training the ResNet. The Wasserstein generative adversarial network was used for generating more new learning samples. At the same time, pretraining model technology was applied to reduce the training parameters, in order to avoid residual neural network overfitting. The results demonstrate that our proposed method achieves a 91.4% accuracy rate, which is better than those of support vector machine, k-nearest neighbor, naïve Bayes model and ensemble learning. In summary, our proposed method demonstrates the potential application of deep learning in pesticide residue detection, expanding the application of THz spectroscopy.

Non-destructive Detection of Insect Foreign Bodies in Finishing Tea Product Based on Terahertz Spectrum and Image

Non-destructive testing of low-density and organic foreign bodies is the main challenge for food safety control. Terahertz time-domain spectroscopy (THz-TDS) and imaging technologies were applied to explore the feasibility of detection for insect foreign bodies in the finishing tea products. THz-TDS of tea leaves and foreign bodies of insects demonstrated significant differences in terms of time domain and frequency signals in the range of 0.3–1.0 THz. These signals were corrected by the use of adaptive iteratively reweighted penalized least squares (AirPLS), asymmetric least squares (AsLS), and baseline estimation and de-noising using sparsity (BEADS) for reducing baseline drift and enhancing effective spectral information. The K-nearest neighbor (KNN) and partial least squares discrimination analysis (PLS-DA) models showed the best performance after AirPLS correction with the prediction accuracy of 98 and 100%, respectively. In addition, the locations and outlines of insect bodies could be clearly presented via the THz-TDS image. These results suggested that THz-TDS spectroscopy and imaging provide an alternative tool for the detection of insect foreign bodies in finishing tea products.

The necessity of periodic boundary conditions for the accurate calculation of crystalline terahertz spectra

Terahertz vibrational spectroscopy has emerged as a powerful spectroscopic technique, providing valuable information regarding long-range interactions - and associated collective dynamics - occurring in solids. However, the terahertz sciences are relatively nascent, and there have been significant advances over the last several decades that have profoundly influenced the interpretation and assignment of experimental terahertz spectra. Specifically, because there do not exist any functional group or material-specific terahertz transitions, it is not possible to interpret experimental spectra without additional analysis, specifically, computational simulations. Over the years simulations utilizing periodic boundary conditions have proven to be most successful for reproducing experimental terahertz dynamics, due to the ability of the calculations to accurately take long-range forces into account. On the other hand, there are numerous reports in the literature that utilize gas phase cluster geometries, to varying levels of apparent success. This perspective will provide a concise introduction into the terahertz sciences, specifically terahertz spectroscopy, followed by an evaluation of gas phase and periodic simulations for the assignment of crystalline terahertz spectra, highlighting potential pitfalls and good practice for future endeavors.

Application of terahertz spectroscopy on monitoring crystallization and isomerization of azobenzene

Terahertz spectroscopy provides a powerful and informative link between infrared spectroscopy and microwave spectroscopy, and is now beginning to make its transition from initial development to broader use by chemists, materials scientists and biologists. In this study, utilizing terahertz spectroscopy we monitored the crystallization and isomerization of azobenzene. In flash-frozen trans-azobenzene solutions, the processes of crystallization and phase transition were observed. A new phase has been experimentally confirmed to exist stably at low temperatures. The results on gradual-frozen experiment indicate that the formation of the observed new phase is determined by the cooling rate. Besides, based on the distinctive spectral features of the isomers, the thermal- and photo-induced isomerization processes of azobenzene were investigated. This work presents that the terahertz spectroscopy has a great potential to study the phase transitions and crystallization of liquid samples under different freezing conditions.