Modified Partial Least Squares Research Articles

Optimizing protein content prediction in rice bean (Vigna umbellata L.) by integrating near-infrared reflectance spectroscopy, MPLS, deep learning, and key wavelengths selection algorithms

Rice bean, a potential protein-rich underutilized crop, can serve as a valuable source for food and feed, contributing to sustainable food and nutrition security. However, traditional methods for measuring protein content are laborious, tedious, and expensive. This study evaluates the performance of Modified Partial Least Squares (MPLS) and 1D Convolutional Neural Network (1D CNN) models in predicting protein content in rice bean using Near-Infrared Reflectance Spectroscopy (NIRS) data. Both models were validated with whole spectra, key wavelengths, and spectral segments derived from Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Competitive Adaptive Reweighted Sampling (CARS). The 1D CNN model demonstrated superior predictive performance, with an 11 % increase in R² and an 18 % increase in RPD using the whole spectra compared to the MPLS model. Specifically, the 1D CNN model with CARS-selected wavelengths achieved an R² of 0.82 and an RPD of 2.15, while the MPLS model achieved an R² of 0.84 and an RPD of 2.25. The segmented spectra approach further enhanced the 1D CNN model's performance, achieving the highest R² of 0.84 and an RPD of 2.28 with CARS-selected segments. These results highlight that the 1D CNN model with CARS-selected wavelengths provides the best predictive performance for protein content in rice bean. The optimized method can lead to rapid early-phase breeding practices and improved quality assessments in leguminous crops, emphasizing the importance of integrating advanced modeling techniques with NIRS for agricultural applications.

Comparative analysis of modified partial least squares regression and hybrid deep learning models for predicting protein content in Perilla (Perilla frutescens L.) seed meal using NIR spectroscopy

Perilla seed meal (PSM), a byproduct of oil extraction from Perilla frutescens L. seeds, is rich in protein (24.26–42.85%) and holds potential as an economical and sustainable animal feed. Traditional methods for assessing protein content are labor-intensive and costly. This study explores Near-Infrared Reflectance Spectroscopy (NIRS) for the rapid, precise, and non-destructive determination of PSM protein content in 126 samples. We developed and evaluated Modified Partial Least Squares (MPLS) regression and deep learning (DL) models, including 1D-CNN (Convolutional Neural Network), LSTM Long Short-Term Memory), and hybrid architectures incorporating skip connections, inception modules, and spectral derivatives. Model performance was validated externally using parameters such as RSQexternal (R-squared), bias, SEP(C) (Standard Error of Prediction), RPD (Residual Prediction Deviation), slope, SD (Standard Deviation), p-value (≥0.05), and the correlation between reference and predicted values. The 1D CNN-LSTM-Inception derivative 1 model achieved the best performance (RPD: 8.0, RSQexternal: 0.98), followed by the MPLS-based model (RPD: 4.88, RSQexternal: 0.96) and the 1D CNN derivative 1 model (RPD: 3.07, RSQexternal: 0.96). These models provide a reliable and advanced technology for the non-destructive screening of PSM protein content, thus aiding in the rapid identification and selection of superior perilla chemotypes from varied backgrounds.

Development and Validation of Near-Infrared Reflectance Spectroscopy Prediction Modeling for the Rapid Estimation of Biochemical Traits in Potato.

Potato is a globally significant crop, crucial for food security and nutrition. Assessing vital nutritional traits is pivotal for enhancing nutritional value. However, traditional wet lab methods for the screening of large germplasms are time- and resource-intensive. To address this challenge, we used near-infrared reflectance spectroscopy (NIRS) for rapid trait estimation in diverse potato germplasms. It employs molecular absorption principles that use near-infrared sections of the electromagnetic spectrum for the precise and rapid determination of biochemical parameters and is non-destructive, enabling trait monitoring without sample compromise. We focused on modified partial least squares (MPLS)-based NIRS prediction models to assess eight key nutritional traits. Various mathematical treatments were executed by permutation and combinations for model calibration. The external validation prediction accuracy was based on the coefficient of determination (RSQexternal), the ratio of performance to deviation (RPD), and the low standard error of performance (SEP). Higher RSQexternal values of 0.937, 0.892, and 0.759 were obtained for protein, dry matter, and total phenols, respectively. Higher RPD values were found for protein (3.982), followed by dry matter (3.041) and total phenolics (2.000), which indicates the excellent predictability of the models. A paired t-test confirmed that the differences between laboratory and predicted values are non-significant. This study presents the first multi-trait NIRS prediction model for Indian potato germplasm. The developed NIRS model effectively predicted the remaining genotypes in this study, demonstrating its broad applicability. This work highlights the rapid screening potential of NIRS for potato germplasm, a valuable tool for identifying trait variations and refining breeding strategies, to ensure sustainable potato production in the face of climate change.

Study on advanced partial least squares for quality-related fault detection

The issue of quality-related fault detection in the industrial process has attracted much attention in recent years. The partial least squares (PLS) is considered an efficient tool for predicting and monitoring. The modified partial least squares (MPLS) is an extended algorithm for solving the oblique decomposition of PLS, however, the study indicated that the loss of quality variable information may affect the prediction of quality information in the decomposition process of the MPLS algorithm. Furthermore, the detection rate of traditional statistics and static control limit is low, and the existing dynamic control limit has certain limitations. Therefore, a new PLS space-decomposition algorithm called advanced partial least squares (APLS) is proposed. APLS avoids the loss of quality information by orthogonal decomposition of process variables according to their relationship with quality. APLS has a more accurate prediction of quality when process variables contain more noise; the fault false alarm rates (FAR) of quality-related faults are reduced by using the new statistics and thresholds combined with local information increment technology in the process variable principal component subspace. Finally, the effectiveness of the proposed approach is verified by a numerical example and an industrial benchmark problem.

A Comparison of Spectral Preprocessing Methods and Their Effects on Nutritional Traits in Cowpea Germplasm

ABSTRACTCowpea (Vigna unguiculata L. (Walp)) is a multipurpose legume, which has good nutritional properties. Nutritional parameters assessed conventionally can be labour intensive, costly and time taking for germplasm screening. Near‐infrared reflectance spectroscopy (NIRS) is a rapid and nondestructive method, which can facilitate high‐throughput germplasm screening. In our study, estimation of amylose and sugars has been done using NIRS. Two preprocessing methods, that is, SNV‐DT (standard normal variate with detrending) and MSC (multiplicative scatter correction), were performed for optimization of the original spectra. Subsequently, MPLS (modified partial least square) regression method was employed to construct the prediction models. In amylose, the best RSQexternal (coefficient of determination) (0.962) was found in SNV‐DT with mathematical treatment 3,8,8,2. The same result was shown in sugar where the best RSQexternal (0.914) was found in SNV‐DT with mathematical treatment 3,4,4,1. Overall, in the case of amylose and sugars, SNV‐DT was found to be a good preprocessing treatment than MSC. Paired t‐test values in all the treatments for both the preprocessing methods were > 0.05 indicating their reliability. High RSQexternal values for both the traits imply the applicability of the prediction models. Thus, these models can facilitate high‐throughput germplasm screening in different national and international crop improvement programmes focusing on quality traits.

Real-Time Monitoring of Fecal Nitrogen Excretion to the Environment Using Near-Infrared Reflectance Spectroscopy: A Preliminary Study in Rabbits

The excessive excretion of nitrogen (N) by farm animals can pose severe environmental risks. In this study, near-infrared reflectance (NIR) spectroscopy (NIRS) was used to explore the feasibility of developing a real-time in situ monitoring tool for fecal N excretion in rabbits. A total of 70 feed and 282 fecal samples from an in vivo digestibility experiment were used. Feed and fecal NIR spectra were employed to develop chemometric models using modified partial least squares (MPLS) for feed and feces, and artificial neural networks (ANN) for feces to predict dietary and fecal N content and fecal N digestibility. Very good accuracy was achieved in predicting feed N (R2val = 0.96; standard error of prediction, SEP = 0.15) and fecal N (R2val = 0.88; SEP = 0.44) content, whereas N digestibility models from wet fecal spectra had a relatively low precision (R2val = 0.70; SEP = 0.018) with MPLS methodology. In contrast, ANNs yielded more robust prediction models for fecal (R2val = 0.98; SEP = 0.25) N content and N digestibility (R2val = 0.91; SEP = 0.012) using wet feces. In conclusion, NIRS calibration with ANNs can be a suitable tool for monitoring the environmental load of N with high precision and accuracy.

Development and validation of near-infrared spectroscopy procedures for prediction of cassava root dry matter and amylose contents in Ugandan cassava germplasm.

Cassava utilization for food and/or industrial products depends on inherent properties of root dry matter content (DMC) and the starch fraction of amylose content (AC). Accordingly, in the present study, near-infrared reflectance spectroscopy (NIRS) models were developed to aid breeding and selection of DMC and AC as critical industrial traits taking care of root sample preparation and cassava germplasm diversity available in Uganda. Upon undertaking calibrations and cross-validations, best models were adopted for validation. DMC in calibration samples ranged from 20 to 45 g 100g-1, whereas, for amylose content, it ranged from 14 to 33 g 100g-1. In the validation set, average DMC was 29.5 g 100g-1, whereas, for amylose content, it was 24.64 g 100g-1. For DMC, a modified partial least square regression model had regression coefficients (R2) of 0.98 and 0.96, respectively, in the calibration and validation set. These were also associated with low bias (-0.018) and ratio of performance deviation that ranged from 4.7 to 5.0. In addition, standard error of prediction values ranged from 0.9 g 100g-1 to 1.06 g 100g-1. For AC, the regression coefficient was 0.91 for the calibration set and 0.94 for the validation set. A bias equivalent to -0.03 and a ratio of performance deviation of 4.23 were observed. These findings confirm the robustness of NIRS in the estimation of dry matter content and amylose content in cassava roots and thus justify its use in routine cassava breeding operations. © 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Prediction models of the nutritional quality of fresh and dry Brachiaria brizantha cv. Piatã grass by near infrared spectroscopy

ABSTRACT This study aimed to generate prediction models to estimate the chemical composition of fresh and dry Brachiaria brizantha cv. Piatã grass using near infrared spectroscopy (NIRS). Chemical analyses of 249 samples were performed to determine oven-dried sample (ODS), dry matter (DM), crude protein (CP), neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent lignin (ADL), cellulose (CEL) and total digestible nutrients (TDN). The samples were scanned in an NIRS spectrometer and different percentages were used to compose and develop the models (100% fresh; 100% dry; 25% fresh:75% dry; 50% fresh:50% dry and 75% fresh:25% dry). The purpose of these mixed models is to know if it is possible to obtain reliable predictions from fresh samples in a database that contains dry samples. The calibration models were developed using modified partial least squares (MPLS) and evaluated by statistical parameters, including coefficient of determination (R²) and residual predictive deviation (RPD). The model with 100% dry samples obtained the best results in R² and RPD validations, for CP (0.94; 3.98), NDF (0.92; 3,49) and TDN (0.90; 3.12). The 100% fresh samples produced the best R² results in ODS (0.83), CP (0.85), ADF (0.84) and ADL (0.83). A screening model was validated to predict the characteristics and components of the fresh samples. The model using 100% dry grass was suitable for predicting all the variables, except ODS, DM and CEL. Highlights Prediction models can be used for assessment of fresh forage, allowing producers to make quicker decisions, thereby saving time and money.

Development of non‐destructive models to predict oil content and fatty acid composition of Gomenzer (Ethiopian mustard) using near‐infrared reflectance spectroscopy

AbstractBackground‟Gomenzer” or Ethiopian mustard is the potential oilseed crop that is economically important. Ethiopian mustard and rapeseed quality breeding has focused on altering the fatty acid contents of seed oil to create novel genotypes with different oil characteristics. The goal of this research was to establish calibration equations' using a method called near‐infrared reflectance spectroscopy with modified partial least squares (MPLS) regression.ResultThe spectra of 180 mustard samples were collected and their oil and fatty acid compositions were determined by n‐hexane extraction and GC–MS methods respectively. With 130 samples, calibration equations were developed for oil and fatty acid compositions. All developed equation had an acceptable value of R 2c, RPDc, and also had suitable 1‐VR values. Prediction of an external validation with 50 datasets revealed a blameless correlation between reference values and NIRS values based on the R 2v in prediction, SEP, and the ratio of SD to SEP (RPDv) for oil, oleic, linoleic, linolenic, and erucic acids. These had suitable values of RPDv and R 2v, with the range of 2.0 to 8.8 and 0.79 to 0.99 respectively.ConclusionThe developed models gave meaningful quantitative data, however, the equations generated for palmitic, stearic, and eicosenoic acids were insufficient for data analysis. Hence, the developed NIRS model, which was satisfactory, could be used to assess and evaluate the oil and fatty acid content for the unknown mustard samples on a regular basis in the oilseed breeding quality program.

Predicting fecal composition, intake, and nutrient digestibility in beef cattle consuming high forage diets using near infrared spectroscopy.

The objective of this study was to develop near infrared spectroscopy (NIRS) calibrations to predict fecal nutrient composition, intake, and diet digestibility from beef cattle fed high forage diets. Heifers were fed 12 different forage-based diets (>95% forage dry matter basis) in 3 total collection digestibility studies, resulting in individual fecal samples and related spectra (n = 135), corresponding nutrient intake, and apparent total tract digestibility (aTTD) data. Fecal samples were also collected from steers grazing two annual and two perennial forage mixtures over two growing seasons. Samples (n = 13/paddock) were composited by paddock resulting in 30 samples from year 1, and 24 from year 2. The grazing fecal spectra (n = 54) were added to the existing fecal composition spectral library. Dried and ground fecal samples were scanned using a FOSS DS2500 scanning monochromator (FOSS, Eden Prairie, MN). Spectra were mathematically treated for detrend and scatter correction and modified partial least squares (MPLS) regression was performed. The coefficient of determination for cross validation (R2cv) and standard error of cross validation (SECV) were used to evaluate the quality of calibrations. Prediction equations were developed for fecal composition [organic matter (OM), nitrogen (N), amylase-treated ash-corrected neutral detergent fiber (aNDFom), acid detergent fiber (ADF), acid detergent lignin (ADL), undigestible NDF after 240 h of in vitro incubation (uNDF), calcium (Ca), and phosphorus (P)], digestibility [DM, OM, aNDFom, N], and intake [DM, OM, aNDFom, N, uNDF]. The calibrations for fecal OM, N, aNDFom, ADF, ADL, uNDF, Ca, P resulted in R2cv between 0.86 and 0.97 and SECV of 1.88, 0.07, 1.70, 1.10, 0.61, 2.00, 0.18, and 0.06, respectively. Equations predicting intake of DM, OM, N, aNDFom, ADL, and uNDF resulted in R2cv values between 0.59 and 0.91, SECV values of 1.12, 1.10, 0.02, 0.69, 0.06, 0.24 kg·d-1, respectively, and SECV values between 0.00 and 0.16 when expressed as % body weight (BW). Digestibility calibrations for DM, OM, aNDFom, and N resulted in R2cv ranging from 0.65 to 0.74 and SECV values from 2.20 to 2.82. We confirm the potential of NIRS to predict fecal chemical composition, digestibility, and intake of cattle fed high forage diets. Future steps include validation of the intake calibration equations for grazing cattle using forage internal marker and modelling energetics of grazing growth performance.

A high throughput phenotyping technique for banana cultivar Sukali Ndizi based on internal fruit quality attributes

Background: Sukali Ndizi quality traits such as Total soluble solid (TSS) content, pulp texture and sugar/acid (S/A) ratio are critical in quality assessment. Screening very large numbers of fruit genotypes has prompted the development of a high throughput method using Near Infrared spectrometry (NIRS). Results: The calibration procedure for the attributes of TSS, pulp texture and S/A ratio was optimized with respect to a reference sampling technique, scan averaging, spectral window, data pre-treatment and regression procedure. Calibration equations for all analytical characteristics were computed by NIR Software ISI Present WINISI using Modified Partial Least Squares (MPLS) and Partial Least Squares. The quality of calibration models were evaluated by Standard Error of Calibration and coefficient of determination parameters between the measured and the predicted values. The results obtained with FOSS NIR systems 2500 spectrometer (model DS 2500) using the 350-2500 nm range, showed good prediction of the quality traits TSS content, pulp texture and S/A ratio. The MPLS method produced satisfactory Calibration model performance for TSS, texture and S/A ratio, with typical Rc2 of 0.73%Brix, 0.69kgf and 0.7; and root mean squared standard error of calibration of 0.73%Brix, 0.25kgf and 5.36 respectively. This is a good set of quality traits predicting Sukali Ndizi quality with NIRS with robustness, as it was obtained by using diverse Ndizi populations. Conclusions: This can be a useful tool to phenotype large numbers of Ndizi hybrids per day, making it possible to reduce on the resources spent when utilizing organoleptic evaluation selection technique.

Estimation of Proximate, Fatty Acid, Mineral Content and Proline Level in Amaranth using Near Infrared Reflectance Spectroscopy

For successful development of new amaranth varieties, it is important to find inexpensive and rapid analysis methods for the measurement of proximate, fatty acid, mineral content, and proline level in seeds. In this study, calibration equations in NIR spectroscopy were developed to estimate for the fatty acid, mineral content and proline level of amaranth using the modified partial least squares (MPLS) regression method. The calibrations estimated by NIR spectroscopy were consistent with the correlations between reference values at external validation. The equations developed were evaluated based on the relative estimate determination results for external validation (RPDv). The equations for total protein (RPDv = 2.967), fat (RPDv = 4.396), Zn (RPDv = 3.668), proline (RPDv = 6.692), oleic acid (RPDv = 3.366) and linoleic acid (RPDv = 2.086) showed high accuracy, while the equations for ash (RPDv = 1.675) and Fe (RPDv = 1.565) showed relatively high accuracy. When calculated with the same validation factors, the level of Ca (RPDv = 0.268), palmitic acid (RPDv = 1.434), stearic acid (RPDv = 0.949), linolenic acid (RPDv = 1.244) and arachidic acid (RPDv = 0.402) were lower than the estimated value. Protein, oil, ash, Fe, Zn, proline, oleic acid and linoleic acid can be used as reliable users, while equations developed for Ca, palmitic acid, stearic acid, linolenic acid and arachidic acid can be reliably used to screen samples for amaranth breeding programmes.

PSVI-6 Predicting Fecal Composition Using Near Infrared Spectroscopy (Nirs): Expanding the Calibration to Include Grazing Beef Samples

Abstract A near-infrared spectroscopy (NIRS) calibration was previously developed to predict fecal composition using samples from beef heifers fed high forage diets ( > 95% forage dry matter basis) during total collection digestibility studies. The objective of the current study was to expand the fecal composition calibration with samples from grazing beef cattle. Fecal samples were collected from beef steers grazing two annual and two perennial forage mixtures over 2 growing seasons. Individual samples (n = 12/paddock) were composited by paddock resulting in 30 samples from year 1, and 24 from year two. Fecal samples were oven dried at 55°C for 48 hours and ground through a 1.0 mm screen prior to scanning on a FOSS DS2500 scanning monochromator (FOSS, Eden Prairie, MN). The grazing fecal spectra (n = 54) was added to the existing library and then mathematically treated for scatter correction. Modified partial least squares (MPLS) regression was performed to develop equations to predict fecal composition [organic matter (OM), nitrogen (N), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), undigestible NDF (uNDF), calcium (Ca), and phosphorus (P)]. The calibrations for fecal OM, N, NDF, ADF, ADL, uNDF, Ca, P resulted in R2CV between 0.86 and 0.96 and SECV of 1.73, 0.07, 1.65, 1.20, 0.63, 1.91, 0.21, and 0.07, respectively. This study confirms the potential of NIRS to predict fecal chemical composition of beef cattle fed high forage or grazing forage diets. Future steps include expansion and further validation of the calibration equations to include digestibility and intake predictions by estimating the internal markers lignin and uNDF.

158 Predicting Intake and Digestibility of Nutrients in Beef Cattle fed High Forage Diets Using Near Infrared Spectroscopy (Nirs) of Feces and Internal Markers

Abstract Objectives of this study were to determine the potential of near infrared-spectroscopy (NIRS) scanning of feces to predict diet digestibility and intake in beef cattle fed high forage diets, and to investigate the use of acid detergent lignin (ADL) and undigestible neutral detergent fiber (uNDF) as internal markers to determine digestibility. Beef heifers were fed 12 different forage-based diets ( > 95% forage dry matter basis) in 3 total collection studies, resulting in individual fecal samples and related spectra (n = 135) corresponding to apparent total tract digestibility (aTTD) and intake data. Dried and ground fecal samples were scanned using a FOSS DS2500 scanning monochromator (FOSS, Eden Prairie, MN). Modified partial least squares (MPLS) regression was performed to develop equations to predict digestibility [dry matter (DM), organic matter (OM), NDF, and nitrogen (N)], and intake [DM, OM, NDF, N, uNDF]. Digestibility prediction equations for DM, OM, NDF, and N resulted in R2CV or 0.72, 0.65, 0.74 and 0.69, respectively, and standard error of cross validation (SECV) between 2.20 to 2.82. Intake calibrations for DM, OM, N, NDF, ADL and uNDF resulted in R2CV values between 0.59 and 0.91, SECV for intake (kg/d) of 1.12, 1.10, 0.02, 0.69, 0.06, 0.24, respectively, SECV for intake % body weight (BW) between 0.00 and 0.16; and SECV for g/kg BW0.75 between 0.60 and 0.86. Significant correlations (P < 0.01) were identified between NDFD determined by aTTD and by the internal marker ADL (r = 0.72) or uNDF (r = 0.57). Developed calibrations predicting marker calculated digestibility resulted in R2CV(SECV) values of 0.85(2.70) and 0.80(2.46) for lignin and uNDF, respectively. We confirm the potential of NIRS to predict digestibility and intake of cattle fed high forage diets. Future steps include further validation of the intake calibration equations using internal marker and energetics modeling.

Application of energy dispersive X-ray fluorescence spectrometry and near-infrared reflectance spectroscopy combined with multivariate statistical analysis for discriminating the geographical origin of soybeans

Energy dispersive X-ray fluorescence spectrometry (ED-XRF) and near-infrared reflectance spectroscopy (NIRS), two non-destructive and rapid techniques, were explored to classify the geographical origin of soybeans. Multivariate statistical methods, including principal component analysis (PCA), discriminant analysis (DA), and modified partial least squares (MPLS), were applied as a chemometric tool to classify soybeans into two groups. A statistically significant difference was observed in the ED-XRF elemental determination of Al, P, S, Cl, Ca, Fe, Ni, Cu, Zn, Br, and Rb between domestic and imported soybeans. A total of 112 samples were almost perfectly classified according to their origin in the application of canonical DA with ED-XRF. NIRS based on DA application showed excellent classification results with 99.1% accuracy after optimizing spectral preprocessing. NIRS based on the MPLS model with leave-one-out cross-validation showed 100% prediction results using first or second derivative pretreatment of the raw spectrum with Salvitzy-Golay or Norris derivative smoothing techniques. The best calibration and validation statistics in the MPLS model for the root mean square error of calibration (RMSEC) and root mean square error of cross-validation (RMSECV) were found to be 0.132% (R2c = 0.9606) and 0.146% (R2cv = 0.9687), respectively. These results suggest that ED-XRF and NIRS combined with multivariate statistical analysis can be suitable technologies for the efficient determination of the geographical origin of soybeans.

Near-Infrared Spectroscopy to Predict Provitamin A Carotenoids Content in Maize

Vitamin A deficiency (VAD) is a public health issue worldwide. Provitamin A (PVA) biofortified maize serves as an alternative to help combat VAD. Breeding efforts to develop maize varieties with high PVA carotenoid content combine molecular and phenotypic selection strategies. The phenotypic assessment of carotenoids is currently done using liquid chromatography, a precise but time- and resource-consuming methodology. Using near-infrared spectroscopy (NIRS) could increase the breeding efficiency. This study used ultra-performance liquid chromatography (UPLC) data from 1857 tropical maize genotypes as a training set and NIRS data to do an independent test of a set of 650 genotypes to predict PVA carotenoids using Bayesian and modified partial least square (MPLS) regression models. Both regression methods produced similar prediction accuracies for the total carotenoids (r2 = 0.75), lutein (r2 = 0.55), zeaxanthin (r2= 0.61), β-carotene (r2 = 0.22) and β-cryptoxanthin (BCX) (r2 = 0.57). These results demonstrate that Bayesian and MPLS regression of BCX on NIRS data can be used to predict BCX content, the current focus on PVA enhancement, and thus offers opportunities for high-throughput phenotyping at a low cost, especially in the early stages of PVA maize breeding pipeline when many genotypes must be screened.

Simultaneous determination of apparent amylose, amylose and amylopectin content and classification of waxy rice using near-infrared spectroscopy (NIRS)

Rice starch properties of apparent amylose content (AAC), amylose content (AC), and amylopectin content (AP) are considered as the most important factors influencing grain quality as they are highly correlated with eating quality. This report is the first effort of predicting AC and AP values in rice flours, and recognizing waxy rice from non-waxy rice using NIRS technique. Calibration models generated by different mathematical, preprocessing treatments and combinations of wavelengths and signals were compared and optimized. The model established by modified partial least squares (MPLS) with “2, 8, 8, 2”/ Inverse MSC and ∼138 wavelengths signals yielded high RSQ of 0.977, 0.928, and 0.912 for AAC, AC and AP, respectively, as simultaneous measurement. MPLS-DA (discriminant analysis) could classify waxy and non-waxy rice with 100% accuracy. This high-throughput technology is valuable for breeding programs, and for the purposes of quality control in the food industry.

Adaptive LII-RMPLS based data-driven process monitoring scheme for quality-relevant fault detection

The partial least squares (PLS) method has been successfully applied for fault diagnosis in industrial production. Compared with the traditional PLS methods, the modified PLS (MPLS) approach is available for slow-time-varying data processing and quality-relevant fault detecting. However, it encounters heavy computational load in model updating, and the static control limits often lead to the low fault detection rate (FDR) or high false alarm rate (FAR). In this article, we first introduce the recursive MPLS (RMPLS) method for quality-relevant fault detection and computational complexity reducing, and then combine the local information increment (LII) method to obtain the time-varying control limits. First, the proposed LII-RMPLS method is capable of quality-relevant faults detection. Second, the adaptive threshold leads to higher FDRs and lower FARs compared with traditional methods. Third, the adaptive parameter-matrices-based model updating approach ensures that the proposed method has better robustness and lower computational complexity when dealing with time-varying factors.

Measurements of Chemical Compositions in Corn Stover and Wheat Straw by Near-Infrared Reflectance Spectroscopy.

Simple SummaryRapid and non-destructive methods play an important role in assessing forage quality. This study is aimed at establishing a calibration model that predicts the moisture, CP, NDF, ADF, and hemicellulose of corn stover and wheat straw by NIRS. In addition, we also intended to compared the predictive accuracy of combined calibration models to the individual models of chemical compositions for corn stover and wheat straw by NIRS. We show that accurately combining calibrated models would be useful for a broad range of end users. Furthermore, the accuracy of the calibration models was improved by increasing the sample numbers (the range of variability) of different straw species.Rapid, non-destructive methods for determining the biochemical composition of straw are crucial in ruminant diets. In this work, ground samples of corn stover (n = 156) and wheat straw (n = 135) were scanned using near-infrared spectroscopy (instrument NIRS DS2500). Samples were divided into two sets, with one set used for calibration (corn stover, n = 126; wheat straw, n = 108) and the remaining set used for validation (corn stover, n = 30; wheat straw, n = 27). Calibration models were developed utilizing modified partial least squares (MPLS) regression with internal cross validation. Concentrations of moisture, crude protein (CP), and neutral detergent fiber (NDF) were successfully predicted in corn stover, and CP and moisture were in wheat straw, but other nutritional components were not predicted accurately when using single-crop samples. All samples were then combined to form new calibration (n = 233) and validation (n = 58) sets comprised of both corn stover and wheat straw. For these combined samples, the CP, NDF, and ADF were predicted successfully; the coefficients of determination for calibration (RSQC) were 0.9625, 0.8349, and 0.8745, with ratios of prediction to deviation (RPD) of 6.872, 2.210, and 2.751, respectively. The acid detergent lignin (ADL) and moisture were classified as moderately useful, with RSQC values of 0.7939 (RPD = 2.259) and 0.8342 (RPD = 1.868), respectively. Although the prediction of hemicellulose was only useful for screening purposes (RSQC = 0.4388, RPD = 1.085), it was concluded that NIRS is a suitable technique to rapidly evaluate the nutritional value of forage crops.

Assessment of Total Fat and Fatty Acids in Walnuts Using Near-Infrared Hyperspectral Imaging

“Persian” walnut (Juglans Regia L.) is one of the most consumed tree nuts in the world. It is rich in several bioactive compounds, with polyunsaturated and monounsaturated fatty acids (PUFA and MUFA) appearing at high concentrations. Walnut consumption protects against cardiovascular, carcinogenic, and neurological disorders. The fatty acid profile has usually been determined by gas chromatography, a reliable and robust tool, but also complex, polluting, and time consuming. In this study, near infrared hyperspectral imaging has been used for the screening of total fat, MUFA, PUFA, saturated, and individual fatty acids in walnuts. Five different walnuts varieties have been considered and modified partial least square (MPLS) regressions have been performed. The SEs of prediction (SEP) in external validation (ranged from 2.12% for PUFA to 13.08% for MUFA) suggest that hyperspectral imaging can be a reliable tool for controlling these parameters in a simple, non-destructive and environmentally friendly way.