Prediction of lamb meat fatty acid composition using near-infrared reflectance spectroscopy (NIRS)

New approach to improve the calibration of main fatty acids by near-infrared reflectance spectroscopy in ruminant meat

This study aims to investigate alternative near-infrared reflectance spectroscopy (NIRS) strategies for predicting beef polyunsaturated fatty acids (PUFA) composition, which have a great nutritional interest, and are actually poorly predicted by NIRS. We compared the results of NIRS models for predicting fatty acids (FA) of beef meat by using two databases: a beef database including 143 beef samples, and a ruminant database including 76 lamb and 143 beef samples. For all the FA, particularly for PUFA, the coefficient of determination of cross-validation (R2CV) and the residual predictive deviation (RPD) of models increased when the ruminant muscle samples database was used instead of the beef muscle database. The R2CV values for the linoleic acid, total conjugated linoleic acid and total PUFA increased from 0.44, 0.79 and 0.59 to 0.68, 0.9, 0.8, respectively, and RPD values for these FA increased from 1.33, 2.14, 1.54 to 1.76, 3.11 and 2.24, respectively. RPD above 2.5 indicates calibration model is considered as acceptable for analytical purposes. The use of a universal equation for ruminant meats to predict FA composition seems to be an encouraging strategy.

A multi-model fusion strategy for multivariate calibration using near and mid-infrared spectra of samples from brewing industry

Near and mid-infrared (NIR/MIR) spectroscopy techniques have gained great acceptance in the industry due to their multiple applications and versatility. However, a success of application often depends heavily on the construction of accurate and stable calibration models. For this purpose, a simple multi-model fusion strategy is proposed. It is actually the combination of Kohonen self-organizing map (KSOM), mutual information (MI) and partial least squares (PLSs) and therefore named as KMICPLS. It works as follows: First, the original training set is fed into a KSOM for unsupervised clustering of samples, on which a series of training subsets are constructed. Thereafter, on each of the training subsets, a MI spectrum is calculated and only the variables with higher MI values than the mean value are retained, based on which a candidate PLS model is constructed. Finally, a fixed number of PLS models are selected to produce a consensus model. Two NIR/MIR spectral datasets from brewing industry are used for experiments. The results confirms its superior performance to two reference algorithms, i.e., the conventional PLS and genetic algorithm-PLS (GAPLS). It can build more accurate and stable calibration models without increasing the complexity, and can be generalized to other NIR/MIR applications.

Improvement of spectral calibration for food analysis through multi-model fusion

Near-infrared (NIR) spectroscopy will present a more promising tool for quantitative analysis if the predictive ability of the calibration model is further improved. To achieve this goal, a new ensemble calibration method based on uninformative variable elimination (UVE)-partial least square (PLS) is proposed, which is named as ensemble PLS (EPLS), meaning a fusion of multiple PLS models. In this method, different calibration sets are first generated by bootstrap and different PLS models are obtained. Then, the UVE is used to shrink the original variable space into a specific subspace. By repeating this process, a fixed number of candidates PLS member models are obtained. Finally, a smaller part of candidate models are integrated to produce an ensemble model. In order to verify the performance of EPLS, three NIR spectral datasets from food industry were used for illustration. Both full-spectrum PLS and UVEPLS of single models were used as reference. It was found that the proposed method could lead to lower RMSEP (root mean square error of prediction) value than PLS and UVEPLS and such an improvement is statistically significant according to a paired t-test. The results showed that the method is of value to enhance the predictive ability of PLS-based calibration involving complex NIR matrices in food analysis.

Application of NIRS for predicting fatty acids in intramuscular fat of rabbit

The aim of this study was to evaluate the use of near infrared reflectance spectroscopy (NIRS) for predicting fatty acid content in intramuscular fat to be applied in rabbit selection programs. One hundred and forty three freeze-dried Longissimus muscles (LM) were scanned by NIRS (1100-2498nm). Modified Partial Least Squares models were obtained. Equations were selected according to standard error of cross validation (SECV) and coefficient of determination of cross validation (R(2)(CV)). Residual predictive deviation of cross validation (RPD(CV)) was also studied. Accurate predictions were reported for IMF (R(2)(CV)=0.98; RPD(CV)=7.57), saturated (R(2)(CV)=0.96; RPD(CV)=5.08) and monounsaturated FA content (R(2)(CV)=0.98; RPD(CV)=6.68). Lower accuracy was obtained for polyunsaturated FA content (R(2)(CV)=0.83; RPD(CV)=2.40). Several individual FA were accurately predicted such as C14:0, C15:0, C16:0, C16:1, C17:0, C18:0, C18:1 n-9, C18:2 n-6 and C18:3 n-3 (R(2)(CV)=0.91-0.97; RPD(CV)>3). Long chain polyunsaturated FA and C18:1 n-7 presented less accurate prediction equations (R(2)(CV)=0.12-0.82; RPD(CV)<3).