Coefficient Of Determination In Calibration Research Articles

Measurement of chemical composition in wet whole maize silage by visible and near infrared reflectance spectroscopy

Visible (Vis) and near infrared reflectance (NIR) spectroscopy were used to predict dry matter (DM), acid and neutral detergent fibre (ADFom and aNDFom), ash, crude protein (CP) and pH in wet whole maize (WWM) silage samples. Samples were analysed by reference methods and spectra collected using a NIR spectrophotometer in reflectance (400–2500 nm). Predictive equations were developed using modified partial least squares (MPLS) with full cross validation using raw spectra or second derivative. Two scatter corrections were also used, namely standard normal variate and detrend (SNVD) and multiplicative scatter correction (MSC). Coefficient of determination in calibration R cal 2 and standard error of cross validation (SECV) were 0.91 (SECV: 6.5 g kg −1 DM) for CP, 0.85 (SECV: 27.4 g kg −1) for DM using SNVD, and 0.86 (SECV: 22.1 g kg −1 DM) for ADFom using raw spectra. Poor correlations between the reference methods and NIR were found for aNDFom, ash, OMD and pH. It was concluded that NIR spectroscopy might be a suitable method to predict DM, CP and ADFom on WWM silage samples with minimal sample preparation.

Quality Determination of Chinese Rice Wine Based on Fourier Transform near Infrared Spectroscopy

To evaluate the applicability of near infrared (NIR) spectroscopy for the determination of five enological parameters (alcoholic degree, pH value, total acid, amino acid nitrogen and °Brix) in Chinese rice wine, transmission spectra were collected in the spectral range from 800 nm to 2500 nm in a 1 mm path length rectangular quartz cuvette with air as the reference at room temperature. Five calibration equations for the enological parameters were established between the reference data and NIR spectra by partial least squares (PLS) regression, separately. The best calibration results were achieved for the determination of alcoholic degree and °Brix. The coefficients of determination in calibration ( R2cal) for alcoholic degree and °Brix were 0.93 and 0.96, respectively. The predictive deviation ratio ( RPD) value of the calibration for alcoholic degree was higher than 3 (3.95), which demonstrated the robustness of the calibration model. The RPD value for °Brix was 2.34. The performance of the calibration models for pH, total acid and amino acid nitrogen were not as good as that of alcoholic degree and °Brix. The RPD values for the three parameters were 1.41, 1.38 and 1.27, respectively and R2cal values were 0.97, 0.83 and 0.90, respectively. In the validation step, r2val values for alcoholic degree and °Brix were all higher than 0.9. The performance of pH and total acid were acceptable [the coefficients of determination in validation ( r2val) for pH and total acid were 0.82 and 0.77, respectively]. The performance of amino acid nitrogen model was the worst, with an r2val value of 0.62. The results demonstrated that the NIR spectroscopic technique could be used to predict the concentrations of these five enological parameters in Chinese rice wine.

Predicting intramuscular fat, moisture and Warner-Bratzler shear force in pork muscle using near infrared reflectance spectroscopy

AbstractPartial least-squares (PLS) models based on visible (Vis) and near infrared reflectance (NIR) spectroscopy data were explored to predict intramuscular fat (IMF), moisture and Warner Bratzler shear force (WBSF) in pork muscles (m. longissimus thoracis) using two sample presentations, namely intact and homogenized. Samples were scanned using a NIR monochromator instrument (NIRSystems 6500, 400 to 2500 nm). Due to the limited number of samples available, calibration models were developed and evaluated using full cross validation. The PLS calibration models developed using homogenized samples and raw spectra yielded a coefficient of determination in calibration (R2) and standard error of cross validation (SECV) for IMF (R2=0·87; SECV=1·8 g/kg), for moisture (R2=0·90; SECV=1·1 g/kg) and for WBSF (R2=0·38; SECV=9·0 N/cm). Intact muscle presentation gave poorer PLS calibration models for IMF and moisture (R2<0·70), however moderate good correlation was found for WBSF (R2=0·64; SECV=8·5 N/cm). Although few samples were used, the results showed the potential of Vis-NIR to predict moisture and IMF using homogenized pork muscles and WBSF in intact samples.

Maturity, Variety and Origin Determination in White Grapes (Vitis Vinifera L.) Using near Infrared Reflectance Technology

Soluble solids content, variety and origin of grapes are three relevant features to determine wine quality. Soluble solids content is directly related to ripeness. The capacity to certify the origin or variety of a grape is of major interest for wine makers. The aim of this paper was to study whether near infrared reflectance (NIR) spectroscopy could be used to determine soluble solids content and to identify different varieties and origins of grapes. Samples of Viura and Chardonnay grapes were collected in two locations (Cadreita and Villamayor de Monjardin) which varied in environmental conditions. Mass and sugars content measured by a refractometer and NIR (800–500 nm) reflectance spectra were obtained. A partial least squares regression (PLS) was made to calibrate the NIR spectra. The regression models for Chardonnay grapes were more robust than for Viura grapes, showing coefficients of determination for calibration and validation of 0.75 and 0.70 and a standard error of cross-validation of 1.27. Therefore, NIR technology is suitable to determine the soluble solids content although it requires a calibration model for each variety, using a large number of data points and variables. The combined variety model showed a residual predictive deviation of 1.33, while Viura and Chardonnay models showed 1.54 and 1.88, respectively. The results of discriminant analysis using variables from the NIR spectrum showed a percentage of well-classified grapes according to the variety to which they belonged of 97.2% and a 79.2% of well-classified Chardonnay grapes according to their origin. In conclusion, NIR technology is a useful alternative for identifying the variety and origin of grapes.

The use of visible (VIS) and near infrared (NIR) reflectance spectroscopy to predict fibre diameter in both clean and greasy wool samples

AbstractAbstract Visible (VIS) and near infrared (NIR) reflectance spectroscopy combined with multivariate data analysis were explored to predict fibre diameter in both clean and greasy Merino wool samples. Fifty clean and 400 greasy wool samples were analysed. Samples were scanned in a large cuvette using a NIRSystems 6500 monochromator instrument by reflectance in the VIS and NIR regions (400 to 2500 nm). Partial least square (PLS) regression was used to develop a number of calibration models between the spectral and reference data. Different mathematical treatments were used during model development. Cross validation was used to assess the performance and avoid overfitting of the models. The NIR calibration models gave a coefficient of determination in calibration (R2) > 0·90 for clean wool samples and a R2 < 0·50 for greasy wool samples. The values for the residual predictive value, RPD (ratio of standard deviation (s. d.) to the root mean square of the standard error of cross validation (RMSECV)) were 3 for clean and 0·6 for greasy wool samples, respectively. The results indicated that fibre diameter in greasy wool samples was poorly predicted with NIR, while clean wool showed good relationships.More research is required to improve the calibration on greasy wool samples if the technology is to be used for rapid analysis to assist in the selection of animals in breeding programmes.

Predicting Intramuscular Fat Content in Pork and Beef by near Infrared Spectroscopy

Prediction ability of near infrared (NIR) spectroscopy for intramuscular fat content (IMF) determination was studied. The material comprised 126 muscle samples; 46 pig longissimus dorsi and semitendinosus and 34 beef longissimus dorsi muscle samples. The IMF content was chemically determined in duplicate using two different chemical methods; fat extraction according to Folch et al. and Soxhlet extraction with hydrolysis according to SIST ISO 1443. Folch extraction underestimated IMF content compared to Soxhlet extraction with hydrolysis (-0.32%, P < 0.0001). Similar repeatability was obtained for Folch and Soxhlet extraction with hydrolysis (0.17% and 0.18%, respectively, P < 0.0001). Sample spectra were scanned from 400–2500 nm by the NIR Systems model 6500 spectrophotometer (Silver Spring, MD, USA) and analysed by WinISI II on minced and intact (pork only) samples. Modified partial least squares regression was used to develop models and to obtain calibration statistics: coefficient of determination in calibration( R2 C) and cross-validation ( R2 CV) and standard error in calibration ( SEC) and cross-validation ( SECV). We prepared different models (for a single muscle/common, by applying NIR spectrum or the whole spectrum, on intact and minced samples). Obtained models proved the remarkable prediction ability of NIR spectroscopy to determine IMF content ( R2 CV between 0.84 and 0.99; SECV between 0.14% and 0.53%) and confirms the potential of NIR spectroscopy to replace laborious chemical procedures. Regarding the factors studied, calibrations were less accurate for intact than for minced samples; the use of an NIR spectrum compared to the whole spectrum had no important effect on the prediction ability. According to calibration statistics, the prediction using a common equation for several muscles seems more reliable than the equations within the muscle, but the latter showed lower bias.

Rapid Prediction of Acid Detergent Fiber, Neutral Detergent Fiber, and Acid Detergent Lignin of Rice Materials by Near-Infrared Spectroscopy

A rapid predictive method based on near-infrared spectroscopy (NIRS) was developed to measure acid detergent fiber (ADF), neutral detergent fiber (NDF), and acid detergent lignin (ADL) of rice stem materials. A total of 207 samples were divided into two subsets, one subset (approximately 136 samples) for calibration and cross-validation and the other subset for independent external validation to evaluate the calibration equations. Different mathematical treatments were applied to obtain the best calibration and validation results. The highest coefficient of determination for calibration (R2) and coefficient of determination for cross-validation (1-VR) were 0.968 and 0.949 for ADF, 0.846 and 0.812 for NDF, and 0.897 and 0.843 for ADL, respectively. Independent external validation still gave a high coefficient of determination for external validation (r2) and a low standard error of performance (SEP) for the three parameters; the best validation results were SEP = 0.933 and r2 = 0.959 for ADF, SEP = 2.228 and r2 = 0.775 for NDF, and SEP = 0.616 and r2 = 0.847 for ADL, indicating that NIR gave a sufficiently accurate prediction of ADF and ADL content of rice material but a less satisfactory prediction for NDF. This study suggested that routine screening for these forage quality parameters with large numbers of samples is possible with NIRS in early-generation selection in rice-breeding programs.

Potential of near-infrared reflectance spectroscopy and chemometrics to predict soil organic carbon fractions

The potential of near-infrared reflectance spectroscopy (NIRS) to predict soil organic C in different particle-size fractions was evaluated. Soil samples ( n = 180) from various crop rotations in Uruguay were analysed by standard chemical and NIRS methods. Partial least squares (PLS) regression with cross validation was used to develop calibrations between reference data and NIRS spectra ( n = 87) and validated using an independent set of samples ( n = 87). Coefficients of determination in calibration ( R CAL 2 ) and standard errors in cross validation (SECV) were 0.90 and 0.6 for coarse-sand C, 0.92 and 0.4 for fine-sand C, and 0.96 and 2.1 for clay + silt C, respectively. Calibrations were poor for C/N ratio ( R CAL 2 < 0.65). Although NIRS demonstrated great potential to predict soil organic C in different particle-size fractions, the nature of sampling and number of samples analysed should be considered in future developments.

Multivariate determination of free fatty acids and moisture in fish oils by partial least-squares regression and near-infrared spectroscopy

The oxidative and hydrolytic degradation of lipids in fish oil was monitored using partial least-squares (PLS) regression and near-infrared reflectance (NIR) spectroscopy. One hundred and sixty ( n=160) fish oil samples from a fishmeal factory were scanned in transflectance by an NIR monochromator instrument (1100–2500 nm). Calibration models were performed for free fatty acids (FFA), moisture (M), peroxide value (PV) and anisidine value (AV). Coefficients of determination in calibration ( R 2) and standard errors of cross validation (SECV) were 0.96 (SECV: 0.59) and 0.94 (SECV: 0.03) for FFA and M in g/kg, respectively. The accuracy of the NIR calibration models were tested using a validation set, yielding coefficients of correlation ( r) and standard errors of prediction (SEP) of 0.98 (SEP: 0.50) and 0.80 (SEP: 0.05) for FFA and M in g/kg, respectively. Poor accuracy ( R 2<0.80) was obtained for the NIR calibration models developed for PV and AV. The paper demonstrates that fish oil hydrolytic degradation of lipids, which seriously affect oil use and storage under industrial conditions, can be successfully monitored using PLS regression and NIR spectroscopy by the fishmeal industry.

Non-destructive prediction of chemical composition in sunflower seeds by near infrared spectroscopy

Near infrared reflectance spectroscopy (NIRS) was explored as a technique to predict moisture (M), oil and crude protein (CP) content on intact sunflower seeds ( Helianthus annuus L.). Three hundred samples were scanned intact in a monochromator instrument NIRS 6500 (NIRSystems, Silver Spring, MD, USA). Calibration equations were developed using modified partial least square regression (MPLS) with internal cross validation. Samples were split in two sets, one set used as calibration ( n=250) where the remaining samples ( n=50) were used as validation set. Two mathematical treatments (first and second derivative), none (log 1/ R) and standard normal variate and detrend (SNVD) as scatter corrections were explored. The coefficient of determination in calibration ( R cal 2) and the standard error in cross validation (SECV) were 0.95 (SECV: 3.3) for M; 0.96 (SECV: 13.1) for CP and 0.90 (SECV: 22.3) for oil in g kg −1 on a dry weight basis (second derivative, 400–2500 nm). Prediction models accounted for less than 65, 70 and 72% of the total variation for oil, M and CP, respectively. However, it was concluded that NIRS is a suitable technique to be used as a tool for rapid pre-screening of quality characteristics on breeding programs.

Exploring the use of near infrared reflectance spectroscopy (NIRS) to predict trace minerals in legumes

The use of near infrared reflectance spectroscopy (NIRS) was explored to predict trace mineral concentrations in two legumes. Samples (332), composite of white clover ( n=97) and lucerne ( n=235), from different locations in Uruguay representing a wide range of soil types, were analysed for sodium (Na), sulphur (S), copper (Cu), iron (Fe), manganese (Mn), zinc (Zn), and boron (B). The samples were scanned in reflectance in a monochromator instrument (400–2500 nm). Calibration models ( n=262) were developed using modified partial least squares regression (MPLS) based on cross-validation and tested using a validation set ( n=70). Two mathematical treatments of the spectra were compared (first and second derivative). The highest coefficients of determination in calibration ( R CAL 2) and the lowest standard errors of cross-validation (SECV) were obtained using second derivative. The R CAL 2 and SECV were 0.83 (SECV: 0.8) for Na and 0.86 (SECV: 2.5) for S in g kg −1 DM; 0.80 (SECV: 4.4), 0.80 (SECV: 10.6), 0.78 (SECV: 22.9), 0.76 (SECV: 0.83) and 0.57 (SECV 25.7) for B, Zn, Mn, Cu and Fe in mg kg −1 DM on a dry weight, respectively. Sulphur (SEP: 5.5), sodium (SEP: 1.2) and boron (SEP 4.2) were well predicted by NIRS on a validation set.

Exploring the Use of near Infrared Reflectance Spectroscopy to Study Physical Properties and Microelements in Soils

Near infrared (NIR) reflectance spectroscopy was used to predict the content of silt, sand, clay, iron (Fe), copper (Cu), manganese (Mn) and zinc (Zn) in soil. A total of 332 samples from agricultural soils (0–15 cm depth) in Uruguay (South America) were used. The samples were scanned in a monochromator instrument (NIRSystems 6500, Silver Spring, MD, USA). Two mathematical treatments (first and second derivative) with SNVD (scatter normal variate and detrend) and without scatter correction were studied. Modified partial least squares (mPLS) was used to develop the calibration models. The coefficient of determination in calibration ( R2cal) and the standard error in calibration ( SEC) using the second derivative were 0.81 ( SEC: 5.1), 0.83 ( SEC: 5.3), 0.92 ( SEC: 2.6) for percent sand, silt and clay, respectively. The R2cal and standard error of cross-validation ( SECV) were for Cu 0.87 ( SEC: 0.7), for Fe 0.92 ( SEC: 21.7), for Mn 0.72 ( SEC: 83.0) and for Zn 0.72 ( SEC: 1.2) on mg kg−1 dry matter. It was concluded that NIR reflectance spectroscopy has a great potential as an analytical method for routine analysis of soil texture, Fe, Zn and Cu due the speed and low cost of analysis.

The use of visible and near-infrared reflectance spectroscopy to predict colour on both intact and homogenised pork muscle

Visible and near-infrared reflectance spectroscopy (VIS/NIRS) was used to predict colour on pork muscle samples. Colour values were measured on 44 pork muscles ( n=44) using a Minolta digital camera to determine CIE L* (lightness), CIE a* (redness) and CIE b* (yellowness). Samples were scanned in the visible and near-infrared region of the spectra on a monochromator instrument (400–2500 nm) on both intact and homogenised presentation to the instrument. Two mathematical treatments (first and second derivative) and two scatter transformations to the spectra were explored, none and standard normal variate and detrend (SNV-D). Predictive (VIS/NIRS) calibrations were developed using a modified partial least-squares regression (mPLS) with internal cross validation. The highest coefficients of determination in calibration ( R 2 cal) and lower standard errors in cross validation (SECV) were found for both CIE L* and a*, in homogenised presentation, while poor calibration statistics were found on intact presentation.

The potential of near-infrared reflectance spectroscopy to analyse soil chemical and physical characteristics

Near-infrared reflectance spectroscopy (NIRS) was used for the analysis of soil samples for silt, sand, clay, calcium (Ca), potassium (K), sodium (Na), magnesium (Mg), copper (Cu) and iron (Fe). A total of 332 samples of different soils from Uruguay (South America) were used. The samples were scanned in a NIRS 6500 (NIRSystems, Silver Spring, MD, USA) in reflectance. Cross validation was applied to avoid overfitting of the models. The coefficient of determination in calibration (R^2_{\rm cal}) and the standard errors in cross validation (SECV) were 0·80 (SECV: 6·8), 0·84 (SECV: 6·0), 0·90 (SECV: 3·6) in per cent for sand, silt and clay respectively. For both macro and microelements the R^2_{\rm cal} and SECV were 0·80 (SECV: 0·1), 0·95 (SECV: 2·9), 0·90 (SECV 0·8), for K, Ca, Mg in g/kg respectively, and 0·86 (SECV: 0·82) and 0·92 (SECV: 25·5) for Cu and Fe in mg/kg. It was concluded that NIRS has a great potential as an analytical method for soil routine analysis due to the speed and low cost of analysis.

Determination of honey quality components by near infrared reflectance spectroscopy

SUMMARYNear infrared reflectance spectroscopy (NIRS) was used to assess the chemical composition of fresh honey samples. Samples (n = 110) from different locations across Uruguay were scanned in transreflectance (0.2 mm path—length) in a NIRS 6500 monochromator in both the visible and near infrared region (400–2500 nm). Prediction equations were developed using modified partial least squares and cross validation was applied to avoid overfitting. Calibrations for water content (g kg−1), pH, electrical conductivity (mS/cm), colour (mm Pfund) and hydroxymethylfurfural (mg kg−1) yielded a coefficient of determination in calibration (R2CAL) and standard error in cross validation of 0.96 (SECV: 31); 0.88 (SECV: 0.21); 0.94 (SECV: 0.07); 0.98 (SECV: 4.7) and 0.67 (SECV: 3.9), respectively. It is concluded that NIRS is a useful method to evaluate chemical composition on fresh honey samples.

Use of near Infrared Reflectance Spectroscopy to Analyse Bovine Faecal Samples

Near infrared (NIR) reflectance spectroscopy was used to predict nitrogen (N), acid detergent fibre (ADF), neutral detergent fibre (NDF) and chromium (Cr) in beef faecal samples. One hundred and twenty faecal samples were scanned in a NIRSystems 6500 monochromator instrument over the wavelength range of 400–2500 nm in reflectance. Calibration equations were developed using modified partial least squares (MPLS) with internal cross validation to avoid overfitting. The coefficient of determination in calibration ( R2cal) and the standard error in cross validation ( SECV) were 0.80 (0.74) for N, 0.92 (12.04) for ADF, 0.86 (13.5) for NDF and 0.56 (0.07) for Cr in g kg−1 dry weight, respectively. Results for validation were 0.78 ( SEP: 0.1) for N, 0.74 ( SEP: 7.5) for ADF, 0.85 ( SEP: 8.5) for NDF and 0.10 (0.09) for Cr in g kg−1 dry weight, respectively.

Visible/near infrared reflectance spectroscopy for predicting composition and tracing system of production of beef muscle

AbstractMuscle chemical analysis and muscle identification both were attempted by using visible and near infrared reflectance spectroscopy (NIRS). Seventy-eight beef muscles (m. longissimus dorsi) from Hereford cattle were used. The samples were scanned in a NIRS monochromator instrument (NIRSystems 6500, Silver Spring, MD, USA) in reflectance mode (log 1/R). Both intact and minced muscle presentation to the instrument were explored. Predictive equations were made using ISI software (Infrasoft International, Port Matilda, PA, USA) and muscle identification was performed by Principal Component Analysis (PCA) and Soft Independent Modelling of Class Analogy (SIMCA). The coefficient of determination in calibration (R2CAL) and standard error in cross validation (SECV) for the intact sample presentation were 009 (SECV: 15·6), 0·89 (SECV: 46·9), 0·48 (SECV: 23·9) for moisture (M), fat and crude protein (CP) on g/kg fresh weight basis respectively. R2CAL and SECV for minced sample presentation were 0·41 (SECV: 161), 0·92 (SECV: 43·4), 0·71 (SECV: 20·5) for M, fat and CP on g/kg fresh weight basis respectively. Qualitative analysis of optical information through PCA and SIMCA analysis showed differences in muscles resulting from two different feeding systems.

The assessment of the chemical composition of fishmeal by near infrared reflectance spectroscopy

The use of near infrared reflectance spectroscopy (NIRS) was investigated as an alternative method for predicting moisture (M), oil, crude protein (CP), ash, salt as NaCl, total volatile nitrogen (TVN) and buffer capacity in fishmeal. The NIRS calibration models were developed using the modified partial least squares (MPLS) regression technique. One thousand and ten (n=1010) fishmeal samples were used to predict chemical composition for quality control in the fishmeal industry. Equations were selected based on the lowest cross validation errors (SECV). The coefficient of determination in calibration (R2) and SECV were 0.93 and 3.9 g kg–1 dry matter (DM); 0.85 and 5.7 g kg–1 DM; 0.92 and 3.7 g kg–1 DM; 0.91 and 4.7 g kg–1 DM; 0.88 and 6.7 g kg–1 DM; 0.94 and 1.8 g kg–1 DM; for M, CP, oil, ash, TVN and NaCl, respectively. It was concluded that NIRS can be used as a method to monitor the quality of fishmeal under industrial conditions.

Effect of Sample Presentation and Animal Muscle Species on the Analysis of Meat by near Infrared Reflectance Spectroscopy

The useful wavelengths in both the visible and the near infrared region as well as two sample presentations (intact and minced) were evaluated to assess moisture (M), crude protein (CP) and intra muscular fat (IMF) in lamb ( n = 300), beef ( n = 100) and chicken ( n = 48) muscle samples. Samples were scanned in reflectance in a NIRSystems 6500 (NIRSystems, Silver Spring, MD, USA). Predictive equations were performed using modified partial least squares (MPLS) with internal cross-validation. The coefficient of determination in calibration ( R2CAL) and the standard error in cross-validation ( SECV) were calculated for each chemical parameter. For moisture, crude protein and fat (each expressed as g kg−1), R2CAL and SECV for beef muscle were 0.98, 0.81 and 0.96, respectively, and SECV was 33.1, 21.8 and 44.8 for beef muscle; for chicken muscle the comparable statistics were 0.99, 0.97 and 0.95 and SECV was 6.9, 2.4 and 33.1; while for lamb muscle R2CAL was 0.76, 0.83 and 0.73 and SECV 10.3, 5.5 and 4.7. It was concluded that the minced presentation is the best way to analyse muscle samples. On the other hand, intact presentation could have a great potential for use in the meat industry, although more research will be needed in order to determine quality attributes on meat samples.

Use of Near-Infrared Reflectance Spectroscopy to Analyze Vitamin Content

Near-infrared reflectance spectroscopy (NIRS) has been used to determine the composition of many products used in animal nutrition. This study sought to determine how best to apply NIRS to analyze the concentration of 11 vitamins in a standard premix. Two methods of regression, stepwise and modified partial least-squares, were used to analyze the results. Stepwise regression showed a confounding effect among various wavelengths and was problematic for use at extreme wavelengths. Modified partial least-squares regression gave more accurate results, as indicated by values of the coefficient of determination in calibration (1.0 for all vitamins) and the coefficient of determination in validation (0.96 to 1.0). Application of NIRS in conjunction with modified partial least-squares regression shows promise as a quick and inexpensive procedure for analysis of vitamin content.