Evaluation of local approaches to obtain accurate near-infrared (NIR) equations for prediction of ingredient composition of compound feeds

Optimisation of the predictive ability of NIR models to estimate nutritional parameters in elephant grass through LOCAL algorithms

Elephant grass is a tropical forage widely used for livestock feed. The analytical techniques traditionally used for its nutritional evaluation are costly and time consuming. Alternatively, Near Infrared Spectroscopy (NIRS) technology has been used as a rapid analysis technique. However, in crops with high variability due to genetic improvement, predictive models quickly lose accuracy and must be recalibrated. The use of non-linear models such as LOCAL calibrations could mitigate these issues, although a number of parameters need to be optimized to obtain accurate results. The objective of this work was to compare the predictive results obtained with global NIRS calibrations and with LOCAL calibrations, paying special attention to the configuration parameters of the models. The results obtained showed that the prediction errors with the LOCAL models were between 1.6 and 17.5 % lower. The best results were obtained in most cases with a low number of selected samples (n = 100-250) and a high number of PLS terms (n = 20). This configuration allows a reduced computation time with high accuracy, becoming a valuable alternative for analytical determinations that require ruminal fluid, which would improve the welfare of the animals by avoiding the need to surgically prepare animals to estimate the nutritional value of the feeds.

Assessing the effect of ingredients variability on the composition of the final complete feed for swine

To characterize the variability of 11 feed ingredients and their impact on the final feed, 728 ingredient samples were collected during 5 months in a feed-plant and were analyzed by near-infrared spectrophotometry (NIRS). Six diets for fattening pigs and gestating sows were formulated using regional information of ingredient chemical composition (reference): LIM, limited; EU, common European; and MULT, multi-ingredient; respectively, including 5, 7, and 10 ingredients. The formulas were replicated 15 times using actual chemical composition (NIRS) from three samples per ingredient and month. This theoretical procedure was validated through small-scale manufacturing 30 LIM-diets, which samples were proximal (PA) and NIRS analyzed for dry matter and crude protein (CP) contents. Those mixtures were also PA analyzed. The ingredients showed coefficient of variation (CV %) higher for crude fiber (CF) (2.6%–18.3%) than CP (2.0%–9.3%). Comparing all diets for all chemical components, variability was reduced when including more ingredients from 0.5%–5.5% to 0.3%–2.6% CV. In most cases, the actual chemical composition of the diets underestimated their reference formula (1.3%–10.8%, CP and CF). A deviation from the targeted diet occurs if variability is not regarded. Therefore, a proper method to predict ingredient composition and nutritional value before use may increase the accuracy of diet formulation between 2% and 10%.

Use of visible and near infrared reflectance spectra to predict lipid peroxidation of light lamb meat and discriminate dam's feeding systems

Measurement of thiobarbituric acid reactive substances (TBARS) is a well-established method for determine lipid oxidation in meat. This assay, however, is time-consuming and generates undesired chemical waste. Dam's milk is the principal source of vitamins and provitamins that delay lipid oxidation of light lamb meat; these compounds are stored in the lamb's muscle tissue. Hence, lamb meat could be used to determine the origin of the dam's diet. The aim of this study is to evaluate Near-infrared reflectance spectroscopy (NIRS) as a tool for determining the lipid peroxidation of light lamb meat and differentiate the meat of light lambs according the diet of their dams during lactation (grazing alfalfa, lucerne, or fed a total mixed ration). NIRS using select wavelengths was able to detect the lipid oxidation of meat (TBARS method). NIRS can detect analytes at concentrations of parts per million. Moreover, the feed diets were discriminated successfully.