RMSEP Values Research Articles

Nondestructively Determining Soluble Solids Content of Blueberries Using Reflection Hyperspectral Imaging Technique

Effectively detecting the quality of blueberries is crucial for ensuring that high-quality products are supplied to the fresh market. This study developed a nondestructive method for determining the soluble solids content (SSC) of blueberry fruit by using a near-infrared hyperspectral imaging technique. The reflection hyperspectral images in the 900–1700 nm waveband range were collected from 480 fresh blueberry samples. An image analysis pipeline was developed to extract the spectrums of blueberries from the hyperspectral images. A regression model for quantifying SSC values was successfully established based on the full range of wavebands, achieving the highest RP2 of 0.8655 and the lowest RMSEP value of 0.4431 °Brix. Furthermore, three variable selection methods, namely the Successive Projections Algorithm (SPA), interval PLS (iPLS), and Genetic Algorithm (GA), were utilized to identify the feature wavebands for modeling. The models calibrated from feature wavebands generated an RMSEP of 0.4643 °Brix, 0.4791 °Brix, and 0.4764 °Brix, as well as the RP2 of 0.8507, 0.8397, and 0.8420 for SPA, iPLS, and GA, respectively. Furthermore, a pseudo-color distribution diagram of the SSC values within blueberries was successfully generated based on established models. This study demonstrated a novel approach for blueberry quality detection and inspection by jointly using hyperspectral imaging and machine learning methodologies. It can serve as a valuable reference for the development of grading equipment systems and portable testing devices for fruit quality assurance.

Simultaneous estimation of prasugrel and aspirin in bulk drugs by chemometric methods

A UV–Vis spectrophotometric method enhanced by chemometric techniques, specifically Principal Component Regression (PCR) and Partial Least Squares (PLS) regression, was developed and validated for the simultaneous quantification of prasugrel (PRA) and aspirin (ASP) in bulk drugs and pharmaceutical formulations. The method demonstrated high accuracy, precision, and robustness, achieving mean recoveries of 100.63% for PRA and 100.08% for ASP with relative standard deviations (RSD) below 3%. Both PCR and PLS models showed excellent predictive capabilities, with RMSEP values of 0.45–0.48 for PRA and 0.78–1.13 for ASP, indicating the models’ reliability. In line with green and white chemistry principles, the method minimizes environmental impact by reducing solvent consumption and waste generation compared to traditional chromatographic methods. The Analytical Eco-Scale score was 84, reflecting excellent compliance with green chemistry standards. The method’s simplicity, low energy consumption, and reduced chemical waste further support its alignment with sustainability goals. However, acetonitrile, a hazardous solvent, was still used in small quantities, and solvent recycling was not implemented, slightly affecting the eco-score. To evaluate the method’s greenness, the RGB12 algorithm was applied, achieving a high score of 94.4%, with the majority of parameters related to reagent consumption, waste production, energy efficiency, and safety scoring optimally. The method’s safety, cost-effectiveness, and minimal environmental footprint make it suitable for routine pharmaceutical analysis, particularly in quality control environments where resource efficiency and sustainability are prioritized. Thus, the developed method offers a sustainable, efficient, and environmentally friendly solution for the simultaneous analysis of prasugrel and aspirin in pharmaceutical formulations, making it a valuable tool for routine analysis in the pharmaceutical industry.

Optimization of a tensile strength prediction model for compacted ribbons using NIR-HIS analysis

The tensile strength (TS) of compacted ribbon is a critical quality attribute in the roller compaction process that impacts the quality of the finished product. This study investigated the use of Near Infrared Hyperspectral Imaging Spectroscopy (NIR-HIS) technology for predicting TS of compacted ribbons, considering the effects of surface curvature, different spectral preprocessing methods, and variable selection methods on a predictive model based on Partial Least Squares regression (PLSr). The spectral preprocessing methods evaluated were Mean Centering (MC) and Standard Normal Variate (SNV). The variable selection methods were Filter by Regression Coefficient (REG), Variable Importance in Projection (VIP), Competitive Adaptive Reweighted Sampling (CARS), and Genetic Algorithm (GA). The results indicated that curved surfaces had no significant impact on the predictive performance of the model (p-value of 0.39 for RMSEP). The PLSr-CARS method, combined with MC spectral preprocessing, was successful in selecting and reducing the number of wavelengths from 182 to 5, as indicated by high values of R2pred and RPD, and a low RMSEP value (0.97, 5.75, and 7.60%, respectively). An MLR model using the 5 wavelengths was also developed, showing similar performance to the PLSr model. Both the MLR and PLSr models demonstrated high predictive accuracy and reliability. These models can perform well even when developed using only a few wavelengths, leading to significant reductions in processing time and measurement costs, making them valuable tools for quality control in the pharmaceutical industry.

Investigation of near infrared and Raman fibre optic process sensors for protein determination in milk protein concentrate

This study investigated the potential of two fibre optic process sensors based on near infrared (NIR) or Raman spectroscopic technology for protein measurement in milk protein concentrate (MPC). Partial least squares (PLS) models were developed using NIR, Raman, and fusion of NIR and Raman spectra data. Calibration models developed were optimized by selecting different spectral pre-treatment methods and spectral regions. Overall, the three optimal models (NIR, Raman, fused NIR and Raman) yielded R2p values >0.9 and RMSEP values in the range of 0.168–0.185 %. The optimised fusion model outperformed all Raman models and had a similar protein prediction accuracy (R2p = 0.911 and RMSEP = 0.178 %) compared to the optimised NIR model (R2p = 0.917 and RMSEP =0.168 %). Results of the study demonstrated that both NIR and Raman process probes can be used as process analytical technology (PAT) tools for inline protein measurements of MPC post membrane filtration.

An innovative variant based on generative adversarial network (GAN): Regression GAN combined with hyperspectral imaging to predict pesticide residue content of Hami melon

The rapid and non-destructive detection of pesticide residues in Hami melons plays substantial importance in protecting consumer health. However, the investment of time and resources needed to procure sample data poses a challenge, often resulting in limited data set and consequently leading insufficient accuracy of the established models. In this study, an innovative variant based on generative adversarial network (GAN) was proposed, named regression GAN (RGAN). It was used to synchronically extend the visible near-infrared (VNIR) and short-wave infrared (SWIR) hyperspectral data and corresponding acetamiprid residue content data of Hami melon. The support vector regression (SVR) and partial least squares regression (PLSR) models were trained using the generated data, and subsequently validate them with real data to assess the reliability of the generated data. In addition, the generated data were added to the real data to extend the dataset. The SVR model based on SWIR-HSI data achieved the optimal performance after data augmentation, yielding the values of Rp2, RMSEP and RPD were 0.8781, 0.6962 and 2.7882, respectively. The RGAN extends the range of GAN applications from classification problems to regression problems. It serves as a valuable reference for the quantitative analysis of chemometrics.

Pulsed Electric Fields (PEF) Treatment Prior to Hot-Air and Microwave Drying of Yellow- and Purple-Fleshed Potatoes

AbstractThis study investigated the impact of pulsed electric fields (PEF) treatment combined with hot-air (HA) and microwave (MW) drying on the quality attributes of yellow and purple potato samples. PEF + HA and PEF + MW treatments showed no significant difference in total phenolic content for yellow potatoes. However, the raw purple sample exhibited significantly higher phenolic content compared to purple PEF1 and PEF3 samples. DPPH antioxidant activity for yellow and purple potatoes ranged from 18.48 to 33.19% with HA drying, and from 21.38 to 39.94% with MW drying. MW drying following PEF treatment was more effective in enhancing antioxidant activity compared to HA drying. However, it was observed that PEF levels had no statistical effect on total monomeric anthocyanin. PEF treatment prior to drying may be recommended in practice as it does not cause structural changes in SEM analysis. According to FTIR results, PEF and drying processes caused some changes in the chemical structure of potato samples. In the PCA results, the untreated purple HA sample exhibited characteristics that were most similar to those of the raw purple potato sample. Conversely, the yellow PEF + MW and untreated yellow with MW samples displayed properties closest to the raw yellow potato samples. According to chemometric analysis, the models exhibit high rCV and rPre values coupled with low RMSEC and RMSEP values and negligible disparities between them, indicating strong correlations between the predicted and reference values.

Performance Comparison of Two Portable Near-infrared Devices for Rapid Authentication of Aceh Aromatic Rice ‘Sigupai’

Sigupai rice, Indonesia local aromatic rice varieties grown in South-West region of Aceh, is highly valued for its fragrance and quality, making it susceptible to adulteration. This study compares the performance of two portable Near-infrared (NIR) devices, SCiO and NeoSpectra, for rapid authentication of Sigupai rice. We evaluated 86 samples for qualitative analysis (i.e. authentic vs adulterated rice) and 44 samples for quantitative analysis (i.e. the level of adulteration). For the qualitative analysis using partial least squares-discriminant analysis (PLS-DA), the best estimation model could differentiate authentic and adulterated samples with an accuracy, sensitivity, specificity, and false positive rates of 89.29%, 92.86%, 85.71% and 14.29% for the NeoSpectra and 97.44%, 100%, 94.87%, and 5.13% for the SCiO, respectively at the validation stage. For quantitative analysis using partial least squares-regression (PLS-R), the best estimation model could estimate the level of adulteration with a coefficient of determination (R²), RMSEP, RPD, and consistency values of 0.92, 1.50%, 5.93 and 100.69% for the NeoSpectra and 0.96, 1.31%, 6.83 and 104.78% for the SCiO. Both portable NIR devices could be used as a rapid analysis tool for the authenticity of Sigupai rice with high accuracy. However, in this study the SCiO device showed a better performance. Keywords: Portable NIR device, Authentication, Aromatic rice, Rapid analysis, Sigupai variety.

Determination of total protein and sugar content in soy-based beverages using variable selection methods applied to ATR-FTIR spectroscopy

This study explores the potential of using ATR-FTIR coupled with partial least squares (PLS) and random forest (RF) for predicting total protein (TP), total sugar (TS), reducing sugars (RS), and non-reducing sugars (NRS) in soy-based beverages (SBBs). Employing variable selection techniques such as interval PLS (iPLS), synergy interval PLS (siPLS), uninformative variable elimination (UVE), ordered predictors selection (OPS), bat algorithm (BA), genetic algorithm (GA), and particle swarm optimization (PSO). The OPS-PLS emerges as the optimal model for TS prediction, yielding an RMSEP of 0.114 wt% and R2p of 0.934. GA-PLS excels in TP, RS, and NRS prediction, achieving RMSEP values of 0.024 wt%, 0.273 wt%, and 0.339 wt%, with corresponding R2p values of 0.880, 0.847, and 0.450. In RF modeling, GA-RF is identified as the superior model for all properties. These findings underscore the potential of regression models and ATR-FTIR as effective and environmentally friendly tools for analyzing SBB composition.

Determination of malathion content in sorghum grains using hyperspectral imaging technology combined with stacked machine learning models

The rapid and precise detection of pesticide residues remains a pressing safety issue in the food industry. In this study, a rapid method for analyzing pesticide residues in sorghum was developed, combining hyperspectral imaging (HSI) technology with stacking ensemble learning (SEL) models. The HSI spectral data were preprocessed using the multivariate scatter correction (MSC) algorithm. The gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine (LGBM), and categorical boosting (CatBoost) algorithms were employed to identify the feature wavelengths with high contributions to the predictive model, and the performances of SEL, GBDT, XGBoost, LGBM, and CatBoost to accurately predict the pesticide residues in sorghum samples were compared. The SEL model constructed using the characteristic wavelength selected by CatBoost has the best predictive performance, with RMSEP, RP2, and RPD values of 0.6940 mg/kg, 0.9798, and 7.029, respectively. The study demonstrated that the combination of HSI and SEL enabled the accurate analysis of pesticide residues in sorghum, providing a reference for the utilization of HSI methods to accurately measure the concentrations of pesticide residues in sorghum and other food products.

Quantitative determination of oxides in cement raw meal based on near-infrared spectroscopy and hybrid feature selection strategy

Determining the content of components in cement raw meal is vital in Portland cement manufacturing. Near-infrared spectroscopy (NIRS)enables rapid, safe, and accurate quantitative analysis of the primary constituents of cement raw meal (CaO, SiO2, Al2O3, and Fe2O3), thereby assisting the cement industry in reducing the time delay in raw meal proportioning control and enhancing the quality of the raw meal. This study employed a hybrid feature selection strategy to identify effective feature points in the NIRS of raw meal samples. Effective feature points from the raw meal were integrated with partial least squares (PLS) to develop a regression model. Initially, NIRS data from raw meal samples were obtained using a Fourier NIR spectrometer. Subsequently, sample set partitioning based on joint X-Y distance (SPXY) was utilized to divide the samples into a calibration set (70 samples) and a validation set (22 samples). Next, A backward interval partial least squares (BiPLS) approach, along with various preprocessing algorithms, was then employed to initially screen the optimal waveband combinations of the NIRS. Within these combinations, effective feature points were further identified using competitive adaptive reweighted sampling (CARS), variable combined population analysis (VCPA) and weighted fusion variable space shrinkage approach (WFVSSA). Finally, we developed and compared PLS models based on the full spectrum and three hybrid feature selection methods. The results demonstrated that.WFVSSA offers a comprehensive assessment of wavelength importance, effectively preserving feature points while eliminating redundant information. BiPLS-WFVSSA-PLS outperformed BiPLS-CARS-PLS and BiPLS-VCPAPLS on the validation set, with R2, RMSEP, and RPD values of 0.8676, 0.1431, and 2.8129 for CaO; 0.9212, 0.1216, and 3.6461 for SiO2; 0.9198, 0.0607, and 3.5618 for Al2O3; and 0.8717, 0.0166, and 2.8614 for Fe2O3, respectively. These findings underscore the superior accuracy and stability of BiPLS-WFVSSA-PLS in determining cement raw meal component contents.

Selective determination of Fe (III) with carbon dots as photoluminescence Probes: Chemometric analysis using Excitation-Emission matrices

The selective and accurate determination of metal ions, crucial in various physiological processes within the human body, is of paramount importance, particularly when these ions, such as Fe(III), coexist with others in the sample. The manuscript addresses the challenge of determining Fe(III) selectively in the presence of interfering species. To achieve this, carbon dots (CDs) are employed as photoluminescence probes due to their biocompatibility and low cytotoxicity. The chemometric analysis of photoluminescence (PL) data was used to overcome selectivity challenges associated with CDs. This study utilized fluorescence excitation-emission matrices (EEMs), considered as second-order data, and applied unfolded partial least squares followed by residual bilinearization (U-PLS/RBL) to account for uncalibrated species.The developed model methodology was assessed using spiked samples containing uncalibrated and interfering species, namely Al(III), Pb(II) and Ni(II), at different concentration levels. The achieved limits of detection (LOD) and quantification (LOQ) at 0.02 and 0.06 mmol/L, respectively, highlight the high sensitivity of the approach. Notably, the application of one RBL component significantly reduces RMSEP values in samples spiked individually with each uncalibrated and interfering species, demonstrating the methodology’s effectiveness in overcoming strong interactions with CDs. Interestingly, this reduction was not observed in test samples containing only Fe(III), where optimal results were achieved without the application of the RBL component.This study introduces a novel approach by formulating a selective methodology for the determination of Fe(III) in the presence of interfering species, employing carbon dots (CDs) as photoluminescence probes. The application of chemometric analysis, specifically U-PLS/RBL, to mitigate selectivity challenges, coupled with the demonstrated sensitivity characterized by low LOD and LOQ values, emphasizes the scientific significance of this research. The results highlight the practical viability of the methodology, particularly in complex samples, contributing valuable insights to the field.

Detection of Anthocyanins in Potatoes Using Micro-Hyperspectral Images Based on Convolutional Neural Networks.

The color potato has the function of both a food and vegetable. The color potato not only contains various amino acids and trace elements needed by the human body but also contains anthocyanins. Anthocyanins have many functions, such as antioxidation, inflammation inhibition, vision improvement, and cancer prevention, so colored potatoes are deeply loved by consumers and have good market prospects. However, at present, the detection of anthocyanin content in color potatoes mainly depends on chemical methods, which are time-consuming and laborious, so it is necessary to study a fast and accurate detection method. In this study, microscopic hyperspectral equipment was used to collect the spectral information of the outer skin and inner skin of potatoes. The original spectrum, pretreatment spectrum, and characteristic spectrum variables of the outer skin and inner skin were predicted by the convolution neural network (CNN) algorithm and partial least squares regression (PLS) algorithm, respectively, and the performance of the model was evaluated by the prediction set correlation coefficient (Rp), prediction set root mean square error (RMSEP), correction set correlation coefficient (Rc), correction set root mean square error (RMSEC), and residual prediction deviation (RPD). The results revealed that the inner skin Raw + CNN model constructed under raw spectral data is optimal with Rc = 0.9508, RMSEC = 0.0374%, Rp = 0.9461, RMSEP = 0.2361% and RPD = 4.4933. The inner skin Savitzky-Golay (SG) + Detrend (DET) + CNN model constructed from pre-processed spectral data is optimal with Rc = 0.9499, RMSEC = 0.0359%, Rp = 0.9439, RMSEP = 0.2384%, RPD = 4.6516. The inner skin DET + competitive adaptive reweighted sampling (CARS) +CNN model constructed from the feature-based spectral data was optimal with Rc = 0.9527, RMSEC = 0.0708%, Rp = 0.9457, RMSEP = 0.2711%, and RPD = 4.1623. It can be seen that the Rp, RMSEP, Rc, RMSEC, and RPD values for modeling the spectral information of the inner skin were higher than those of the outer skin under the three different spectral data. The prediction accuracy of the model built by the CNN algorithm was better than the conventional algorithm PLS, the application of the CNN algorithm in inner skin can achieve accurate prediction of anthocyanin content in potato.

Firefly Interval Selection Combined With Extreme Learning Machine for Spectral Quantification of Complex Samples

ABSTRACTFirefly algorithm (FA) combined with extreme learning machine (ELM) is developed for spectral interval selection and quantitative analysis of complex samples. The method firstly segments the spectra into a certain number of intervals. Vectors with 1 and 0, which represent the interval selected or not, are used as the inputs of the FA. The RMSEP value predicted by ELM model is used as the fitness function of the FA. The activation function and number of hidden layer nodes of ELM, number of spectral intervals, population number, environmental absorbance, and constant of FA are optimized. The predictive performance of FA‐ELM is compared with full‐spectrum PLS, ELM, genetic algorithm‐ELM (GA‐ELM), and particle swarm optimization‐ELM (PSO‐ELM) by one ultraviolet (UV) spectrum dataset of gasoil and three near‐infrared (NIR) spectral datasets of corn, wheat, and tablet samples. The results show that FA‐ELM has a better performance compared with its competitors in predicting monoaromatics, water, wheat kernel texture, and active pharmaceutical ingredients (APIs) in gasoil, corn, wheat, and tablet samples.

Diverse local calibration approaches for chemometric predictive analysis of large near-infrared spectroscopy (NIRS) multi-product datasets

This study aimed to assess the predictive accuracy of Near-Infrared Spectroscopy (NIRS) across a large multi-product library, employing novel local calibration methodologies. Three local strategies were examined: LOCAL Algorithm, Locally Weighted Regression predicted on k-nearest neighbor selection (kNN-LWPLSR), along with a newly proposed algorithm within this study called Hybrid Local. These strategies were applied to an extensive multi-product dataset. When compared with Global PLS models, the results exhibited significant reductions in RMSEP values for all local strategies. Particularly, the kNN-LWPLSR demonstrated proficient prediction for the constituents of ADF and DM. The newly proposed method [Hybrid Local] exhibits comparable performance to the LOCAL Algorithm; however, it notably reduces the prediction time by half compared to the latter, representing a significant advancement for the practical implementation of NIRS technology within industrial processing scenarios.

Rapid quantitative analysis of multiple rare earth elements in NdFeB alloys based on laser-induced breakdown spectroscopy (LIBS) and random forest (RF)

NdFeB material has excellent comprehensive magnetic properties, which plays a crucial role in the field of rare earth magnetic materials, and is one of the important basic materials to support modern high-tech industries. In the production of NdFeB alloys, the control of rare earth elements (REEs) is directly related to the quality and resource utilization efficiency of NdFeB materials. Therefore, the prediction of REEs content in NdFeB alloys is of great research significance and application value for the quality control of products, the enhancement of production efficiency and reduction of energy consumption by the rare earth material industry. In this study, a combination of laser-induced breakdown spectroscopy (LIBS) and random forest (RF) was used to investigate the quantitative analysis of four REEs (Nd, Pr, Tb and Dy) in NdFeB alloys. Firstly, the original LIBS spectra were screened by principal component analysis-mahalanobis distance (PCA-MD). Then, the effects of different data processing methods on the screened LIBS spectra were explored, and next, the feature variables were extracted from the preprocessed spectral data by the variable importance measurement (VIM). In order to further verify the prediction performance of the model, the prediction results of the RF models based on the different methods were compared. Finally, a PCA-VIM-RF calibration model was established on the basis of the optimized spectra, selected feature variables and parameters. Leave-one-out cross validation (LOOCV) was used to optimize the parameters of PCA-MD method, spectral preprocessing method and variable importance thresholds during the construction of the calibration model. The results show that the PCA-VIM-RF model has better prediction performance than the RF calibration model based on the raw spectra. For the PCA-VIM-RF calibration method of Nd, Pr, Tb and Dy elements, the values of R2CV are 0.9991, 0.9998, 0.9986, and 0.9984, respectively, the values of RMSECV are 0.06296%, 0.02788%, 0.04647% and 0.05252%, respectively. The values of R2P are 0.9508, 0.9975, 0.9691 and 0.9457, respectively, and the values of RMSEP are 0.6082%, 0.09205%, 0.5776% and 0.2631%, respectively.The above results indicate that LIBS combined with PCA-VIM-RF algorithm is a promising method for rapid quantitative analysis of REEs in NdFeB alloys without complicated sample preparation, which can provide some new ideas or strategies for the future research, development and quality control of rare earth materials.

Online Detection of Dry Matter in Potatoes Based on Visible Near-Infrared Transmission Spectroscopy Combined with 1D-CNN

More efficient resource utilization and increased crop utilization rate are needed to address the growing demand for food. The efficient quality testing of key agricultural products such as potatoes, especially the rapid testing of key nutritional indicators, has become an important strategy for ensuring their quality and safety. In this study, visible and near infrared (Vis/NIR) transmittance spectroscopy (600–900 nm) was used for the online analysis of multiple quality parameters in potatoes. The study concentrated on comparing three one-dimensional convolutional neural network (1D-CNN) models, specifically, the fine-tuned DeepSpectra, the fine-tuned 1D-AlexNet, and classic CNN, with UVE-PLS (uninformative variable elimination–partial least squares) models. These models utilized spectral data for the real-time detection of dry matter (DM) content in potatoes. To address the challenges posed by limited data from Vis/NIR, this study strategically implemented data augmentation techniques. This approach significantly enhanced the robustness and generalization capabilities of the models. The 1D-AlexNet and DeepSpectra models achieved 0.934 and 0.913 R2P and 0.0603 and 0.0695 g/100 g RMSEP for DM, respectively. Compared to UVE-PLS, the R2P value improved by 21.31% (0.770 to 0.934) for the 1D-AlexNet model and 18.64% (0.770 to 0.913) for the DeepSpectra model. The RMSEP value was reduced by 47.31% (0.114 to 0.0603) for 1D-AlexNet, and 39.30% (0.114 to 0.0695) for the DeepSpectra model. As a result, this study would be helpful for researching the online Vis/NIR transmission determination of potato DM using deep learning. These results highlighted the immense potential of employing specific spectral features in deep-learning models for a more precise and efficient online assessment of agricultural quality. This advancement provided some insight and reference for further contributing to the evolution of more targeted and efficient quality assessment methods in agricultural products.

The Employment of ATR-FTIR Spectroscopy and Chemometrics for Authentication of Bawal (Colossoma macromopum) Fish Oil from Palm Oil

Fish is a product with high nutritional content, and the development of fish oil has made it easier for consumers to consume. High-quality fish oil can be a target for adulteration, such as with vegetable oils like palm oil. FTIR spectroscopy was developed for the authentication of bawal fish oil (BFO) from palm oil (PO). Bawal fish oil and palm oil were prepared in binary mixtures with concentrations from 0-100%, resulting in 21 mixtures. The oil was directly measured by attenuated total reflectance (ATR) spectral measurement in the mid-infrared region (4000-600 cm-1). The results were combined with linear discriminant analysis (LDA) and multivariate calibration, such as PLSR or PCR. The results showed that LDA could make clear discrimination between bawal fish oil, palm oil, and their mixtures without any misclassification observed. Multivariate calibration with PLSR using the first derivative spectrum provided the best model for the relationship between actual and predicted FTIR values. At this conditions, the R2 values for the calibration and validation models obtained were 0.9966 and 0.9969, with RMSEC and RMSEP values of 0.0249 and 0.0248, respectively. Therefore, FTIR spectroscopy combined with LDA and PLSR is an effective method for authenticating bawal fish oil from palm oil.

Using Macroscopic, Microscopic, and FTIR Spectroscopy combined with Chemometrics to Authenticate Arabica Coffee from Antbush

Limited coffee stocks, high consumption, export demands, and increasing prices may contribute to counterfeiting. In other countries, including India and Africa, coffee forgery using antbush has been extensively established. Due to its abundance in Indonesia, unscrupulous coffee growers allegedly employ antbush (Senna occidentalis) as an adulterant in coffee to increase commercial earnings. This study aims to authenticate coffee from antbush using macroscopic and microscopic differences and FTIR spectroscopy combined with chemometrics. The arabica coffee samples from various regions were oven-roasted to a second crack and milled. The materials were then examined under macroscopic, microscopic, and infrared spectroscopic conditions. The obtained responses were used to monitor the qualitative and quantitative information in the targeted samples. Antbush in coffee samples is successfully identified microscopically by remnants of palisade tissue as well as the structural differences of the endosperm. In addition, FTIR spectroscopy combined with multivariate calibration can accurately estimate the concentration of antbush as adulterants in the target sample. Principle component regression (PCR) provides the best modeling for the relationship between the actual value. FTIR predicted the value of antbush with the lowest RMSEC and RMSEP values of 0.852 and 0.896, respectively, with the coefficient of determination (R2) in calibration and validation models of 0.9996 and 0.9967, respectively. The combination of macroscopic, microscopic, and FTIR spectroscopy offered reliable tools to authenticate arabica coffee from antbush.

Development of near-infrared spectroscopy calibration model and monitoring software: For monitoring hexamethylenetetramine concentration in hexamethylenetetramine–acetic acid solution

In recent years, near-infrared spectroscopy has been increasingly used in chemical engineering processes. However, accurately and in real-time monitoring concentrations remains highly challenging. This study presents a method for accurately monitoring the hexamethylenetetramine concentration in hexamethylenetetramine-acetic acid solution by integrating near-infrared spectroscopy with software techniques. Compared to the methods of zero mean standardization (Z-score), standard normal variate (SNV), first derivative (D1), second derivative (D2), and vector normalization (VN), the Savitzky-Golay first derivative (SGFD) method effectively addresses issues related to baseline drift, noise, and other interferences. This results in clearer identification of characteristic absorption peaks of different components and a significant improvement in the accuracy of the analysis. The study combined the uninformative variable elimination (UVE), differential evolution (DE), and moth flame optimization (MFO) algorithms to optimize a feature wavenumber selection method, which reduced the number of variables by 78.74 %, and the values of root mean square error of calibration set (RMSEC) and root mean square error of prediction (RMSEP) by 18.86 % and 30.38 %, respectively. Additionally, the analytical method was integrated with real-time monitoring software, enabling the monitoring of hexamethylenetetramine concentration. This integration supports the improvement of the quality and safety of the manufacturing process. The results demonstrate the potential application of near-infrared spectroscopy in real-time monitoring and automated control of chemical processes in the chemical industry.

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.