Assessing Calibrations: SEP, RPD, RER and R2

Fearn, Tom

doi:10.1255/nirn.689

Cited by 286 publications

(218 citation statements)

References 1 publication

Supporting

Mentioning

207

Contrasting

Unclassified

Order By: Relevance

“…An RPD value greater than three was considered adequate for analytical purposes in most of NIR applications for agricultural products (Williams, 2001;Fearn, 2002).…”

Section: Wool Samplesmentioning

confidence: 99%

Feasibility study on the potential of visible and near infrared reflectance spectroscopy to measure alpaca fibre characteristics

Gishen

Cozzolino

2007

Animal

View full text Add to dashboard Cite

Visible (Vis) and near infrared (NIR) reflectance spectroscopy is a rapid and non-destructive technique that has found many applications in assessing the quality of agricultural commodities, including wool. In this study, Vis and NIR spectroscopy combined with multivariate data analysis was investigated regarding its feasibility in predicting a range of fibre characteristics in raw alpaca wool samples. Mid-side samples (n 5 149) were taken from alpacas from a range of colours and ages at shearing time over 4 years (2000 to 2004) and subsequently analysed for fibre characteristics such as mean fibre diameter (MFD) and standard deviation (and coefficient of variation), spin fineness, curvature degree (and standard deviation), comfort factor, medullation percentage (by weight and number in white samples only) using traditional reference laboratory testing methods. Samples were scanned in a large cuvette using a FOSS NIRSystems 6500 monochromator instrument in reflectance mode in the Vis and NIR regions (400 to 2500 nm). Partial least squares (PLS) regression was used to develop a number of calibration models between the spectral and reference data. Mathematical pre-treatment of the spectra (second derivative) as well as various combinations of wavelength range were used in model development. The best calibration model was found when using the NIR region (1100 to 2500 nm) for the prediction of MFD, which had a coefficient of determination in crossvalidation (R 2 ) of 0.88 with a root mean square standard error of cross validation (RMSECV) of 2.62 mm. The results show the NIR technique to have promise as a semi-quantitative method for screening purposes. The lack of grease in alpaca wool samples suggests that the technique might find ready application as a rapid measurement technique for preliminary classing of shorn fleeces or, if used directly on the animal, the technology might offer an objective tool to assist in the selection of animals in breeding programmes or shows.

show abstract

“…An RPD value greater than three was considered adequate for analytical purposes in most of NIR applications for agricultural products (Williams, 2001;Fearn, 2002).…”

Section: Wool Samplesmentioning

confidence: 99%

Feasibility study on the potential of visible and near infrared reflectance spectroscopy to measure alpaca fibre characteristics

Gishen

Cozzolino

2007

Animal

View full text Add to dashboard Cite

show abstract

“…Similarly except raw data and the baseline corrected data models of reflectance spectra gave RPD values of more than 3.0 and SEP as well as RMSEP values of less than 1.0. Since the RPD values in the range of 3.1 to 4.9 are good for screening purposes [16] [17] (Fearn, 2002;Williams, 2001), models that show more than 3.0 were considered as good models. Baseline and MSC corrected reflectance spectra with 1 st derivative model gave the highest RPD value of 4.46 and R 2 of 0.95, also its SEP (0.670) and RMSEP (0.782) are less than the other models with RPD more than 3.0.…”

Section: Resultsmentioning

confidence: 99%

“…RPD values of 1 or less indicate that the model equation predicts the same values in random chances only. RPD values in the range of 3.1 to 4.9 are good for screening purposes and values in the range 5 to 6.4 are good for quality control purposes [16] [17] (Fearn, 2002;Williams, 2001). Bias values, which correspond to the average difference between the standard reference and predicted values, were also considered in model selection.…”

Section: Discussionmentioning

confidence: 99%

Application of NIR Reflectance Spectroscopy on Rapid Determination of Moisture Content of Wood Pellets

Sundaram¹,

Mani²,

Kandala³

et al. 2015

AJAC

View full text Add to dashboard Cite

NIR spectroscopy was used to measure the moisture concentration of wood pellets. Pellets were conditioned to various moisture levels between 0.63% and 14.16% (wet basis) and the moisture concentration was verified using a standard oven method. Samples from various moisture levels were separated into two groups, as calibration and validation sets. NIR absorption spectral data from 400 nm to 2500 nm with 0.5 nm intervals were collected using pellets within the calibration and validation sample sets. Spectral wavelength ranges were taken as independent variables and the MC of the pellets as the dependent variable for the analysis. Measurements were obtained on 30 replicates within each moisture level. Partial Least Square (PLS) analysis was performed on both raw and preprocessed spectral data of calibration set to determine the best calibration model based on Standard Error of Calibration (SEC) and coefficient of multiple determinations (R 2 ). The PLS model that yielded the best fit was used to predict the moisture concentration of validation group pellets. Relative Percent Deviation (RPD) and Standard Error of Prediction (SEP) were calculated to validate goodness of fit of the prediction model. Baseline and Multiple Scatter Corrected (MSC) reflectance spectra with 1 st derivative model gave the highest RPD value of 4.46 and R 2 of 0.95. Also it's SEP (0.670) and RMSEP (0.782) were less than the other models those had RPD value more than 3.0 with less number of factors. Therefore, this model was selected as the best model for moisture content prediction of wood pellets.

show abstract

“…Such analysis requires a simple yet sturdy model that is capable of maintaining its predictive capability over a prolonged period, coupled with instrumentation that is similarly robust in terms of operational lifetime. The capability of the calibration model to successfully predict unknown samples (i.e., samples not present in the calibration set used to construct the model) must also be assessed; this is done by applying the model to a small number of samples for which the models target property for prediction is defined [3,33,[50][51][52][53][54][55]. Once the model's results are comparable with those of the reference values, the model can be considered to be accurate and useful for determining that target property in the future analysis of unknown samples.…”

Section: Chemometricsmentioning

confidence: 99%

“…An assessment of the model's accuracy is essential to avoid overfitting; consequently, different validation procedures should be applied, as a calibration model without validation is nonsense. In feasibility studies, cross-validation is a practical method to demonstrate that the instrumental method can predict something; however, the predictive ability of the method needs to be demonstrated using an independent validation set [3,33,[50][51][52][53][54][55][56][57][58].…”

Section: Chemometricsmentioning

confidence: 99%

The Application of State-of-the-Art Analytic Tools (Biosensors and Spectroscopy) in Beverage and Food Fermentation Process Monitoring

et al. 2017

View full text Add to dashboard Cite

Abstract:The production of several agricultural products and foods are linked with fermentation. Traditional methods used to control and monitor the quality of the products and processes are based on the use of simple chemical analysis. However, these methods are time-consuming and do not provide sufficient relevant information to guarantee the chemical changes during the process. Commonly used methods applied in the agriculture and food industries to monitor fermentation are those based on simple or single-point sensors, where only one parameter is measured (e.g., temperature or density). These sensors are used several times per day and are often the only source of data available from which the conditions and rate of fermentation are monitored. In the modern food industry, an ideal method to control and monitor the fermentation process should enable a direct, rapid, precise, and accurate determination of several target compounds, with minimal to no sample preparation or reagent consumption. Here, state-of-the-art advancements in both the application of sensors and analytical tools to monitor beverage and food fermentation processes will be discussed.

show abstract

Assessing Calibrations: SEP, RPD, RER and R²

Cited by 286 publications

References 1 publication

Feasibility study on the potential of visible and near infrared reflectance spectroscopy to measure alpaca fibre characteristics

Feasibility study on the potential of visible and near infrared reflectance spectroscopy to measure alpaca fibre characteristics

Application of NIR Reflectance Spectroscopy on Rapid Determination of Moisture Content of Wood Pellets

The Application of State-of-the-Art Analytic Tools (Biosensors and Spectroscopy) in Beverage and Food Fermentation Process Monitoring

Contact Info

Product

Resources

About