Ratio Of Prediction To Deviation Research Articles

Chemometrics for estimating the fermentation and quality properties of kimchi based on hyperspectral image analysis

Kimchi is a traditional Korean dish made from fermenting vegetables. The fermentation process is crucial for enhancing its quality and flavor during storage. Approaches such as hyperspectral imaging (HSI) and chemometrics (PLS, partial least square; SVR, support vector regression) including principal component analysis (PCA), and 2-dimensional correlation spectroscopy (2D-COS) can detect key physical and chemical components and changes in total soluble solids (TSS), pH, titratable acidity (TA), salinity, and lactic acid bacteria (LAB). Multivariate analytical models were developed to predict the quality properties using full and characteristic wavelengths and preprocessed data. The results showed that the ratio of prediction to deviation (RPD) values of the PLS prediction model constructed using the full wavelengths of TSS, salinity, pH, TA, and LAB were 1.57, 2.33, 2.79, 2.91, and 2.73, respectively. The Savitzky Golay 1st derivative preprocessed SVR model established based on characteristic wavelengths (951, 1020, 1139, 1174, 1216, 1321, and 1384 nm) extracted by PCA and a 2D-COS matrix showed the best results and increased efficiency in predicting pH (Rp2 = 0.9166, RPD = 3.281) and the number of LAB (Rp2 = 0.8488, RPD = 2.466). Additionally, the visualization process accurately illustrated the distribution of various quality indicators of kimchi across different periods. These results demonstrate that our proposed HSI strategy successfully assessed the degree of kimchi fermentation.

An optimized spectral reconstruction method for shift excitation Raman differential spectroscopy

Raman spectroscopy is a powerful analytical method, but when the composition of the test sample is intricate, the original spectral data may contain noise and fluorescence background interference, making it more difficult to extract Raman spectral information from the original spectra. Especially the fluorescence background signal, which is typically several orders of magnitude stronger than the Raman signal, can even overwhelm or obscure the Raman signals, thereby impeding the qualitative or quantitative analysis of the Raman spectra. One effective method for removing the fluorescence background is shift excitation Raman differential spectroscopy (SERDS), which typically involves measuring two raw Raman spectra using slightly different excitation wavelengths, combined with reconstruction algorithms, to obtain Raman spectra free from fluorescence interference. For this purpose, a reconstruction method based on Tikhonov regularized least squares (TRLS) was developed in this study, which mitigated the oscillations caused by the direct unconstrained least squares (DULS) reconstruction method. The method was verified and optimized using four groups of artificial datasets with different characteristics. By selecting an appropriate value for parameter α, the relative standard deviation (RSD) of the reconstructed datasets was lower than that of the artificial datasets in most cases. Additionally, we evaluated the performance of the TRLS reconstruction algorithm based on a quantitative model of real Raman spectral datasets, assessing the algorithm’s performance from three perspectives: the root mean square error (RMSE), the correlation coefficient (R), and the ratio of prediction to deviation (RPD). The quantitative results indicate that using the TRLS method for reconstruction enhances both prediction accuracy and practicality. In summary, findings from both simulated data and actual experiments demonstrate that the TRLS-based reconstruction method substantially improves the stability and reliability of differential Raman spectra reconstruction.

Using visible and near infrared spectroscopy and machine learning for estimating total petroleum hydrocarbons in contaminated soils

Petroleum pollution in soil is very damaging to the areas affected by the accidental release of petroleum hydrocarbons and has destructive impacts on natural resources and environmental health. Therefore, its monitoring and analysis are critical, however, due to the cost and time associated with chemical approaches, finding a quick and cost-effective analytical method is valuable. This study was conducted to evaluate the potential of using visible near infrared (Vis-NIR) spectroscopy to predict total petroleum hydrocarbons (TPH) in polluted soils around the Shadegan ponds, in southern Iran. One hundred soil samples showing various degrees of pollution were randomly collected from topsoil (0–10 cm). The soil samples were analyzed for TPH using Vis-NIR reflectance spectroscopy in the laboratory and then following application of preprocessing transformation, partial least squares PLS regression as well as two machine learning models including random forest (RF), and support vector machine (SVM) were examined. The results showed that the reflectance values at 1725 nm and 2311 nm, respectively, served as indicative TPH reflectance features, exhibiting weaker reflection with rising TPH. Among the preprocessing methods, the baseline correction method indicated the highest performance than the others. According to the evaluation model criteria in the validation dataset, the efficiency of the three selected models was observed in the following order SVM > RF > PLS regression. The SVM model provided the best performance in the validation dataset with r2 = 0.85, root mean of square (RMSEP = 1.59 %, and the ratio of prediction to deviation (RPD) = 2.6. Overall, this study provided strong evidence supporting the considerable potential of Visible-NIR spectroscopy as a rapid and cost-effective technique for estimating TPH levels in oil-contaminated soils, surpassing traditional chemical analytical methods. Applying the mid-infrared spectrum (MIR) in combination with Visible-NIR data is expected to provide more comprehensive and accurate results when assessing soils in polluted areas, thereby enhancing the accuracy and reliability of the results across a diverse range of soil types.

Detection of moisture content of edamame based on the fusion of reflectance and transmittance spectra of hyperspectral imaging

AbstractEdamame is a nutritious and economically valuable soybean. The moisture content is an important indicator of the quality of the edamame. The traditional methods in the detection of moisture content of edamame have the disadvantage of large detection errors. In this research, the fusion of transmittance and reflectance spectra of hyperspectral imaging combined with chemometrics was proposed to predict the moisture content of edamame. Also, the effect of different preprocessing of the spectra on the predictive performance was analyzed. Single spectra, primary fusion spectra, and intermediate fusion spectra were established as the prediction models for partial least squares regression (PLSR) and partial least squares support vector regression (LSSVR), respectively. The results of the prediction models showed that the spectral transform absorption (STA) combined with PLSR has the best prediction performance for a single spectrum with predictive correlation (RP) of 0.7749 and ratio of prediction to deviation (RPD) of 1.7. Standard normal variate (SNV) combined with LSSVR has the best prediction performance for primary fusion spectra with RP of 0.8821 and RPD of 1.9. SNV combined with LSSVR has the best prediction performance for intermediate fusion spectra with RP of 0.9149 and RPD of 2.4. The Rp and RPD of prediction models of the moisture content of edamame based on fusion spectra were significantly improved compared with single spectra. Compared with primary fusion, intermediate fusion is a more suitable fusion strategy. This research provides experimental basis for the prediction of moisture content of edamame using spectral fusion combined with chemometrics.

Influences of season on cattle slurry compositions predicted by micro-NIR spectrometry

In this study, we used a low-cost micro-near-infrared (NIR) spectrometer to analyze the composition of cattle slurries and examined the influence of the season (spring vs. winter) and characteristic wavelength on the performance of the corresponding quantitative models. One hundred and twenty-seven cattle slurry samples collected from five farms in spring and winter (Heilongjiang, China) were characterized in terms of total nitrogen (TN), total phosphorus (TP), total potassium (TK), available phosphorus (AP), available potassium (AK), organic matter (OM), and pH using reference methods. The partial least squares (PLS) method was used to establish seasonal models that involved winter and a mixed model to quantify cattle slurry composition. The characteristic wavelengths were selected using a competitive adaptive reweighted-sampling (CARS) algorithm. The results showed that the winter model successfully predicted the TP content using the full band. The coefficient of determination for prediction (R2p) and ratio of prediction to deviation (RPD) were 0.846 and 3.077, respectively. This model was useful for predicting TK and AK contents and pH (R2p = 0.551–0.773, RPD = 2.251–2.727) and moderately useful for predicting TN and OM contents (R2p = 0.517 and 0.498, respectively; RPD = 1.892 and 2.094, respectively). For all compositions, the prediction performance was better for the winter models than for the mixed models. Although the use of characteristic wavelengths improved the efficiency of model construction, it had no positive effect on the performance of the mixed models. This study enables the setup of a rapid testing method for safely returning slurry to the field.

Application of fourier transform infrared photoacoustic spectroscopy for quantification of nutrient contents and their plant availability in manure and digestate

In this study, we assess the feasibility of using Fourier Transform Infrared Photoacoustic Spectroscopy (FTIR-PAS) to predict macro- and micro-nutrients in a diverse set of manures and digestates. Furthermore, the prediction capabilities of FTIR-PAS were assessed using a novel error tolerance-based interval method in view of the accuracy required for application in agricultural practices. Partial Least-Squares Regression (PLSR) was used to correlate the FTIR-PAS spectra with nutrient contents. The prediction results were then assessed with conventional assessment methods (root mean square error (RMSE), coefficient of determination R2, and the ratio of prediction to deviation (RPD)). The results show the potential of FTIR-PAS to be used as a rapid analysis technique, with promising prediction results (R2 > 0.91 and RPD >2.5) for all elements except for bicarbonate-extractable P, K, and NH4+-N (0.8 < R2 < 0.9 and 2 < RPD <2.5). The results for nitrogen and phosphorus were further evaluated using the proposed error tolerance-based interval method. The probability of prediction for nitrogen within the allowed limit is calculated to be 94.6 % and for phosphorus 83.8 %. The proposed error tolerance-based interval method provides a better measure to decide if the FTIR-PAS in its current state could be used to meet the required accuracy in agriculture for the quantification of nutrient content in manure and digestate.

Simultaneous Rapid Detection of Multiple Physicochemical Properties of Jet Fuel Using Near-Infrared Spectroscopy.

Jet fuel is the primary fuel used in the aviation industry, and its quality has a direct impact on the safety and operational efficiency of aircraft. The accurate quantitative detection and analysis of various physicochemical property indicators are important for improving and ensuring the quality of jet fuel in the domestic market. This study used near-infrared (NIR) spectroscopy to establish a suitable model for the simultaneous and rapid detection of multiple physicochemical properties in jet fuel. Using more than 40 different sources of jet fuel, a rapid detection model was established by optimizing the spectral processing methods. The measurement models were separately built using the partial least-squares (PLS) and orthogonal PLS algorithms, and the model parameters were optimized. The results show that after the Savitzky-Golay second derivative preprocessing, the PLS model built using the feature spectra selected by the uninformative variable elimination wavelength algorithm achieved the best measurement performance. Compared with the PLS model without preprocessing, the range of the resulting accuracy improvement was at least 15.01%. Under the optimal model parameters, the calibration set regression coefficient (Rc2) of the 11 jet fuel property index models ranged from 0.9102 to 0.9763, with the root-mean-square error of calibration values up to 0.8468 °C (for flash points). The regression coefficient (Rp2) of the validation set ranged from 0.8239 to 0.9557, with the root-mean-square error of prediction values up to 1.1354 °C (for flash points). The ratios of prediction to deviation (RPD) values were all in the range of 1.9-3.0, indicating high accuracy and reliability of the model. The rapid NIR analysis method established in this study enables the simultaneous and rapid detection of multiple physicochemical properties of jet fuel, thereby providing effective technical support for ensuring the quality of jet fuel in the market.

Application of Multivariate Regression and Artificial Neural Network Modelling for Prediction of Physicochemical Properties of Grape-Skin Compost.

The reusability of by-products in the food industry is consistent with sustainable and greener production; therefore, the aim of this paper was to evaluate the applicability of multiple linear regression (MLR), piecewise linear regression (PLR) and artificial neural network models (ANN) to the prediction of grape-skin compost's physicochemical properties (moisture, dry matter, organic matter, ash content, carbon content, nitrogen content, C/N ratio, total colour change of compost samples, pH, conductivity, total dissolved solids and total colour change of compost extract samples) during in-vessel composting based on the initial composting conditions (air-flow rate, moisture content and day of sampling). Based on the coefficient of determination for prediction, the adjusted coefficient of determination for calibration, the root-mean-square error of prediction (RMSEP), the standard error of prediction (SEP), the ratio of prediction to deviation (RPD) and the ratio of the error range (RER), it can be concluded that all developed MLR and PLR models are acceptable for process screening. Furthermore, the ANN model developed for predicting moisture and dry-matter content can be used for quality control (RER >11). The obtained results show the great potential of multivariate modelling for analysis of the physicochemical properties of compost during composting, confirming the high applicability of modelling in greener production processes.

Effect of Combined Non-Wood and Wood Spectra of Biomass Chips on Rapid Prediction of Ultimate Analysis Parameters Using near Infrared Spectroscopy

The ultimate analysis parameters, including carbon (C), hydrogen (H), nitrogen (N), and oxygen (O) content in biomass, were rarely found to be predicted by non-destructive tests to date. In this research, we developed partial least squares regression (PLSR) models to predict the ultimate analysis parameters of chip biomass using near-infrared (NIR) raw spectra of non-wood and wood samples from fast-growing tree and agricultural residue and nine different traditional spectral preprocessing techniques. These techniques include first derivative (sd1), second derivative (sd2), constant offset, standard normal variate (SNV), multiplicative scatter correction (MSC), vector normalization, min-max normalization, mean centering, sd1 + vector normalization, and sd1 + MSC. Additionally, we employed a genetic algorithm (GA), successive projection algorithm (SPA), multi-preprocessing (MP) 5-range, and MP 3-range to develop a PLSR model for rapid prediction. A dataset consisting of 120 chip biomass samples was utilized for model development in which the samples were non-wood samples of 65–67% and wood samples of 33–35%, and the model performance was evaluated and compared. The selection of the optimum performing model was mainly based on criteria such as the coefficient of determination in the prediction set (R2P), root mean square error of the prediction set (RMSEP), and the ratio of prediction to deviation (RPD). The optimal model for weight percentage (wt.%) of C was obtained using GA–PLSR, yielding R2P, RMSEP, and RPD values of 0.6954, 1.1252 wt.%, and 1.8, respectively. Similarly, for wt.% of O, the most effective model was obtained using the multi-preprocessing PLSR–5 range method with R2P of 0.7150, RMSEP of 1.3088 wt.%, and RPD of 1.9. For wt.% of N, the optimal model was obtained using the MP PLSR-3 range method, resulting in R2P, RMSEP, and RPD values of 0.6073, 0.1008 wt.%, and 1.6, respectively. However, wt.% of the H model provided R2P, RMSEP, and RPD values of 0.5162, 0.2322 wt.%, and 1.5, respectively. Notably, the limit of quantification (LOQ) values for C, H, and O were lower than the minimum reference values used during model development, indicating a high level of sensitivity. However, the LOQ for N exceeded the minimum reference value, implying the samples to be predicted by the model must be in the range of reference range in the calibration set. By scatter plot analysis, the effect of combined non-wood and wood spectra of biomass chips on rapid prediction of ultimate analysis parameters using NIR spectroscopy was investigated. To include different species in a model, the species have to be not only in the different values of the constituents to make a wider range for a robust model, but also must provide their trend line characteristics in the scatter plot, i.e., correlation coefficient (R), slope, and intercept (same slope and slope approached to 1, and intercept is same (no gap) and approached zero, high R approached to 1). The effect of the R, slope, and intercept to obtain the better-optimized model was studied. The results show that the different species affected the model performance of each parameter prediction in a different manner, and by scatter plot analysis, which of these species were affecting the model negatively and how the model could be improved was indicated. This is the first time the effect has been studied by the principle of a scatter plot.

Assessment of drying characteristics for whole and sliced red ginseng using short‐wave infrared spectroscopy and mathematical model

AbstractMonitoring the moisture content is crucial for red ginseng quality control. In this study, we established a moisture content prediction model using partial least squares regression (PLSR) based on short‐wave infrared (SWIR) data of different ginseng types (whole and sliced) at various drying temperatures and times. The model parameters, including the R2 values, ratio of prediction deviation, and range error ratio, showed that the developed model was robust. Sliced ginseng showed a constant rate period, followed by a falling rate period, whereas whole ginseng showed only a falling rate period. The falling‐rate period coincided with the formation of a dry zone on the sample surface in the moisture distribution map, indicating that monitoring the moisture distribution map can predict the onset of the falling‐rate period. These results suggest that the SWIR technique is a promising nondestructive method for evaluating the drying process of red ginseng.Practical applicationsMonitoring the moisture content of red ginseng is essential to prevent microbial spoilage during transportation. Traditional monitoring of water content is destructive and time‐consuming. Therefore, using hyperspectral imaging, we aimed to develop an efficient moisture prediction model for the drying process of red ginseng. We believe that our study makes a significant contribution to non‐destructive and rapid quality evaluation in red ginseng production.

Determination of Multicomponents in Rubi Fructus by Near-Infrared Spectroscopy Technique

Objective. Rubi Fructus (RF) is a dry aggregate fruit of Rubus (Rosaceae). It has shown significant pharmacological effects such as anti-oxidation, hypoglycemic, and anti-inflammatory. A combination of near-infrared (NIR) spectroscopy and partial least squares regression (PLSR) under seven different spectral data preprocessing techniques was used to determine the performance of quantitative analysis correction models which employed moisure, ellagic acid, and total flavonoids as indicators of RF. Methods. Ninety-seven different RF batches were collected for NIR spectra. By using primary analysis techniques such as drying method, high-performance liquid chromatography (HPLC), and ultraviolet visible spectrophotometry (UV-Vis), the contents of moisure, ellagic acid, and total flavonoids were determined. The NIR spectral data and the primary analysis method data were correlated through PLSR. Seven methods were used for pretreating the spectral data, including no spectral pretreatment, first derivative, standard normalized variate, multiple scattering corrections, elimination of constant offset, and minimum maximum normalization. The quantitative analysis correction models adopted PLSR chemometrics for moisture, ellagic acid and total flavonoids were developed, and their effectiveness was evaluated using the correlation coefficient (R), ratio of prediction to deviation (RPD), and root mean square error (RMSE). Results. The first derivative was combined with variable standardization, elimination of constant offset, and multiple scattering corrections, respectively, to pretreat the PLSR models for moisture, ellagic acid, and total flavonoids. The R-values of the PLSR models for moisture, ellagic acid, and total flavonoids were, respectively, 0.9788, 0.9468, and 0.9748, all of which were higher than 0.90, and the RPD values were 4.9, 3.1, and 4.5, respectively, which were all larger than 3.0. The RMSE ratios of the calibration set and the test set were 0.98, 0.94, and 1.0, respectively. Conclusion. The R-values of the NIR-PLSR models for moisture, ellagic acid, and total flavonoids are all greater than 0.90 after suitable pretreatments, indicating that the models are reliable. The RPD values are more than 3.0, which indicate that the models are good and useable for quality control. The RMSE ratios are closed to 1, indicating that the calibration set and test set had same distribution and the models were not overfitting indicating good predictability.

High-throughput phenotyping of nutritional quality components in sweet potato roots by near-infrared spectroscopy and chemometrics methods

The lack of an efficient approach for quality evaluation of sweet potatoes significantly hinders progress in quality breeding. Therefore, this study aimed to establish a near-infrared spectroscopy (NIRS) assay for high-throughput analysis of sweet potato root quality, including total starch, amylose, amylopectin, the ratio of amylopectin to amylose, soluble sugar, crude protein, total flavonoid content, and total phenolic content. A total of 125 representative samples were utilized and a dual-optimized strategy (optimization of sample subset partitioning and variable selection) was applied to NIRS modeling. Eight optimal equations were developed with an excellent coefficient of determination for the calibration (R2C) at 0.95–0.99, cross-validation (R2CV) at 0.93–0.98, external validation (R2V) at 0.89–0.96, and the ratio of prediction to deviation (RPD) at 6.33–11.35. Overall, these NIRS models provide a feasible approach for high-throughput analysis of root quality and permit large-scale screening of elite germplasm in future sweet potato breeding.

A high-throughput phenotyping method for sugarcane rind penetrometer resistance and breaking force characterization by near-infrared spectroscopy

BackgroundSugarcane (Saccharum spp.) is the core crop for sugar and bioethanol production over the world. A major problem in sugarcane production is stalk lodging due to weak mechanical strength. Rind penetrometer resistance (RPR) and breaking force are two kinds of regular parameters for mechanical strength characterization. However, due to the lack of efficient methods for determining RPR and breaking force in sugarcane, genetic approaches for improving these traits are generally limited. This study was designed to use near-infrared spectroscopy (NIRS) calibration assay to accurately assess mechanical strength on a high-throughput basis for the first time.ResultsBased on well-established laboratory measurements of sugarcane stalk internodes collected in the years 2019 and 2020, considerable variations in RPR and breaking force were observed in the stalk internodes. Following a standard NIRS calibration process, two online models were obtained with a high coefficient of determination (R2) and the ratio of prediction to deviation (RPD) values during calibration, internal cross-validation, and external validation. Remarkably, the equation for RPR exhibited R2 and RPD values as high as 0.997 and 17.70, as well as showing relatively low root mean square error values at 0.44 N mm−2 during global modeling, demonstrating excellent predictive performance.ConclusionsThis study delivered a successful attempt for rapid and precise prediction of rind penetrometer resistance and breaking force in sugarcane stalk by NIRS assay. These established models can be used to improve phenotyping jobs for sugarcane germplasm on a large scale.

Turning Zr(IV) into a phosphate ester mimetic enzyme via de novo synthesis for hydrolyzing organophosphorus warfare agents and rapid activity prediction by near-infrared spectroscopy

Chemical warfare agents, especially organophosphorus warfare agents are a variety of various chemical substances employed in warfare that can massively cause damage to the organism even death. The partial catalytic function of Universitetet i Oslo-66 (UiO-66, a zirconium-based metal–organic framework) produced via de novo synthesis is similar to calf intestinal alkaline phosphatase (CIAP, EC 3.1.3.1) and has a lower usage cost, is expected to be a mimetic enzyme for hydrolyzing organophosphorus warfare agents due to the fact that it has lattice defects. 0.1–5.6 mL of acetic acid was added as a regulator to form different lattice defects. Typically, mimetic enzyme activity tests often require expensive reagents as substrates, and the experiment processes are time-consuming and inconvenient. As a result, a novel approach based on near-infrared spectroscopy (NIRs) was developed in this research for predicting UiO-66 mimetic enzyme activity. p-Nitrophenyl phosphate disodium salt (pNPP) was selected as a lower toxic organophosphorus warfare agent to conduct the subsequent simulant research. The absorbance of the mimetic enzyme catalytic system was evaluated using Ultraviolet–Visible (UV–Vis) spectroscopy to investigate the amount of its activity and compare it with CIAP activity. The synthesized samples’ enzyme activities ranged from 1.91 to 6.67 U/g and their catalytic costs ranged from 0.65 to 2.26 CNY. The powdered samples' near-infrared spectra were obtained. The sample’s enzyme activity data were linked with each sample’s original NIR spectra to establish the NIR model. A partial least squares (PLS) model of mimetic enzyme activity was developed by coupling spectral preprocessing with a variable screening technique. The root-mean-square error of cross-validation (RMSECV), the correlation coefficient of validation set (R) value, and the ratio of prediction to deviation (RPD) value were employed as assessment model indicators. The near-infrared spectrum model was developed by merging the best 2nd derivative spectral preprocessing with the Competitive Adaptive Reweighted Sampling (CARS) variable screening method. This model's RMSECV was 0.41 U/g, the correlation coefficient of calibration set (R_cv) value was 0.916, the root-mean-square error of prediction (RMSEP) set was 0.52 U/g, R was 0.948, and RPD was 2.51. As a result, the established model could be implemented to quantify the activity of UiO-66 quickly by including more variations of de novo-synthesized samples. The prediction method is simple, rapid, low cost and adaptable to be the theoretical and practical basis for further studying other interdisciplinary research work in biomimetic enzymology and spectroscopy.

Development of new fruit quality indices through aggregation of fruit quality parameters and their predictions using near-infrared spectroscopy

This study introduces a flexible approach to aggregate several quality-related parameters to provide new fruit quality indices (FQIs) with the aim of evaluating the overall quality of fruits. The calculations were based on the aggregation of selected fruit quality parameters, such as total soluble solids (TSS), titratable acidity (TA), pH, firmness, dry matter (DM), and lipid content. The fruit quality index 1 (FQI1) and fruit quality index 2 (FQI2) were calculated, respectively, as geometric and weighted arithmetic means of the scaled fruit quality parameters. Both calculations resulted in numerical scores corresponding to a quality rank ranging from 0 to 100. For rapid and non-destructive detection, near-infrared (NIR) spectra were calibrated for predictions of the developed indices using partial least squares (PLS). The predictive performance was evaluated based on several metrics, including root mean square errors of calibration (RMSEC) and prediction (RMSEP), coefficients of determination for calibration (R2) and prediction (Q2), the ratio of RMSEP and RMSEC (RP/C) and the ratio of prediction to deviation (RPD). Mango, tangerine, and avocado were used in the demonstration. The FQI evaluations offer flexibility in the number of aggregate parameters, making them applicable to various types of fruits. With the NIR detection, the developed indices resulted in a significant reduction in the measurements of quality-related parameters but still provided the overall quality of the fruit samples.

Rapid prediction method of ZIF-8 immobilized Candida rugosa lipase activity by near-infrared spectroscopy

Candida rugosa lipase (CRL, EC3.1.1.3) is one of the main enzymes synthesizing esters, and ZIF-8 was chosen as an immobilization carrier for lipase. Enzyme activity testing often requires expensive reagents as substrates, and the experiment processes are time-consuming and inconvenient. As a result, a novel approach based on near-infrared spectroscopy (NIRs) was developed for predicting CRL/ZIF-8 enzyme activity. The absorbance of the immobilized enzyme catalytic system was evaluated using UV–Vis spectroscopy to investigate the amount of CRL/ZIF-8 enzyme activity. The powdered samples' near-infrared spectra were obtained. The sample’s enzyme activity data were linked with each sample’s original NIR spectra to establish the NIR model. A partial least squares (PLS) model of immobilized enzyme activity was developed by coupling spectral preprocessing with a variable screening technique. The experiments were completed within 48 h to eliminate inaccuracies between the reduction in enzyme activity with increasing laying-aside time throughout the test and the NIRs modeling. The root-mean-square error of cross-validation (RMSECV), the correlation coefficient of validation set (R) value, and the ratio of prediction to deviation (RPD) value were employed as assessment model indicators. The near-infrared spectrum model was developed by merging the best 2nd derivative spectral preprocessing with the Competitive Adaptive Reweighted Sampling (CARS) variable screening method. This model's root-mean-square error of cross-validation (RMSECV) was 0.368 U/g, the correlation coefficient of calibration set (R_cv) value was 0.943, the root-mean-square error of prediction (RMSEP) set was 0.414 U/g, the correlation coefficient of validation set (R) value was 0.952, and the ratio of prediction to deviation (RPD) was 3.0. The model demonstrates that the fitting relationship between the predicted and the reference enzyme activity value of the NIRs is satisfactory. The findings revealed a strong relationship between NIRs and CRL/ZIF-8 enzyme activity. As a result, the established model could be implemented to quantify the enzyme activity of CRL/ZIF-8 quickly by including more variations of natural samples. The prediction method is simple, rapid, and adaptable to be the theoretical and practical basis for further studying other interdisciplinary research work in enzymology and spectroscopy.

Non-destructive detection of polysaccharides and moisture in Ganoderma lucidum using near-infrared spectroscopy and machine learning algorithm

Ganoderma lucidum, the fruiting body of the Poraceae fungi G. lucidum or Ganoderma sinense, has a long history of use in promoting health and longevity in Asian countries. However, traditional methods for detecting polysaccharides and moisture in G. lucidum are complicated, time-consuming, and damaging (to the sample). In this study, rapid and nondestructive near-infrared (NIR) spectroscopy (700–2500 nm) was uesd to directly scan the back of the G. lucidum cap without powdering. Thereafter, we used synergy interval partial least squares to select the performing band and the ant lion optimization (ALO) algorithm to optimize the least squares support vector machine (LSSVM) model for these two components. The results showed that the ALO-LSSVM model could predict the total polysaccharide and moisture content with high accuracy. The correlation coefficient for calibration were both >0.9 and their ratio of prediction to deviation (RPD) values of prediction were 2.6 and 3.6, respectively, indicating the non-destructive determination of polysaccharides and moisture in G. lucidum will provide great convenience for on-site testing by procurement personnel. This shows the combination of NIR spectroscopy and the ALO-LSSVM algorithm has potential applications for the rapid and nondestructive analysis of natural products.

An automatic generation of pre-processing strategy combined with machine learning multivariate analysis for NIR spectral data

Pre-processing near-infrared (NIR) spectral data is indispensable in multivariate analysis, since the measured spectra of complex samples are often subject to overwhelming background, light scattering, varying noises, and other unexpected factors. Various pre-processing methods have been developed to remove or reduce the interference of these effects. Until now, most applications of NIR spectra pre-processing in multivariate calibration have been trial-and-error, with selecting a proper method depending on the nature of the data, expertise, and practitioner experience. Thus, it is usually challenging to determine the best pre-processing method for a given data. In order to tackle these problems, this study proposes a new concept of data pre-processing, namely, automatically generating a pre-processing strategy (AGoES). This concept belongs to the ensemble pre-processing method, where machine learning algorithms (PLSR, SVM, k-NN, DT, AB, and GPR) built on differently preprocessed data are combined by 5-fold cross-validation and grid search optimization. To investigate our concept, a public NIR spectral dataset was used to predict three responses, including dry matter content (DM), organic matter content (OM) and ammonium nitrogen content (AN) from manure organic waste. The results show that SVM is the best algorithm combined with the AGoES pre-processing to predict DM and AN with a ratio of prediction to deviation (RPD) of 3.619 and 2.996, respectively. The AB tandem with AGoES pre-processing is the best strategy for predicting OM with an RPD of 3.185. Therefore, in the framework of the AGoES concept, it is unsupervised pre-processing, more simple, and feasible to apply multivariate analysis using machine learning algorithms.

Near-Infrared Spectroscopy with Machine Learning for Classifying and Quantifying Nutmeg Adulteration

Near-infrared spectroscopy (NIRS) provides broadbands, overtones, and combinations of organic-bond vibrations and has been used to characterize agricultural and food products. The adulteration of grated nutmeg with cinnamon is extremely profitable and difficult to detect; to prevent retail fraud, it is vital to differentiate between these materials. This study proposes a model for classifying the adulteration of nutmeg with cinnamon and predicting the level of adulteration. NIR spectra were characterized with six machine learning (ML) algorithms, namely, the principal component-multilayer perceptron (PC-MLP), principal component-linear discriminant analysis (PC-LDA), partial least squares regression (PLSR), support vector machine (SVM), random forest (RF), and decision tree (DT) methods. PC-MLP provided 100% accuracy in calibration and prediction in distinguishing nutmeg from cinnamon. In addition, this approach showed excellent performance in predicting the adulteration ratio of nutmeg and cinnamon with a high coefficient of determination of prediction (R 2 pred) value of 0.9969, low root mean square error of prediction (RMSEP) value of 0.5728%, and high ratio of prediction to deviation (RPD) value of 17.9605. Therefore, this study indicates the potential of integrating NIR spectroscopy with PC-MLP to classify and quantify the adulteration of nutmeg.

Nondestructive detection of water status and distribution in corn kernels during hot air drying using multispectral imaging.

The characteristics of corn kernels are strongly connected with the content of three statuses of water: bound water, immobilized water, and free water. Monitoring different water contents is very important to optimize the drying process, improve corn quality, and reduce energy consumption. The feasibility of nondestructive detection of water status and its distribution in corn kernels during the hot-air drying process using multispectral imaging was investigated. The chemometric methods used to develop prediction models were back propagation neural network, least-squares support vector machine, and partial least squares. The back propagation neural network achieved the best prediction performance for total and free water contents, with correlation coefficient of prediction (Rp ) of 0.9717 and 0.9782 respectively, root-mean-square error of prediction (RMSEP) of 4.48% and 2.54% respectively, and ratio of prediction to deviation (RPD) of 4.87 and 4.29 respectively. And partial least squares was better for the prediction of immobilized and bound water contents, with Rp of 0.9612 and 0.9798 respectively, RMSEP of 0.57% and 0.06% respectively, and RPD of 4.78 and 4.42 respectively. It could be concluded that multispectral imaging combined with chemometric methods would be a promising technique for rapid and nondestructive detection of water status and its distribution in corn kernels. © 2023 Society of Chemical Industry.