Submitted to: Food Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/4/2007
Publication Date: 7/20/2007
Citation: Kim, Y., Singh, M., Kays, S.E. 2007. Near-infrared Spectroscopic Analysis of Macronutrients and Energy in Homogenized Meals. Food Chemistry. Volume 105: pages 1248-1255, 2007. Interpretive Summary: Near-infrared (NIR) spectroscopy is a multiple component analytical technique enabling several determinations to be made simultaneously without requiring extensive sample preparation. Near infrared spectroscopy is a technique that measures the amount of light energy reflected by a substance in a specific region of the electromagnetic spectrum and relates the energy to a measured component of the substance by mathematical modeling. Since its use for the analysis of protein and oil content in corn and soybeans, NIR spectroscopy has proved to be an excellent method for nutrient analysis of foods. Compared with the traditional chemical methods for nutrient measurements, NIR spectroscopy is a rapid, environmentally benign method, which has found widespread use in analyses for components such as fat, protein, moisture, and carbohydrate. Based on the success of NIR spectroscopy for rapid measurement of total dietary fiber in mixed meals, this study was conducted to evaluate the feasibility of NIR spectroscopy for the prediction of major components (moisture, crude protein, crude fat, and total carbohydrate) and energy simultaneously in mixed meals without the need for sample preparation beyond homogenization. Performing NIR spectroscopic analysis with the various mixed meals reduced the individual analysis time for each component from at least 1 day for conventional analysis to less than 30 minute. The prediction accuracy was highest for protein, followed by energy, moisture, carbohydrate, and fat. The results show that NIR spectroscopy is an adequate technique for the screening of samples for major components simultaneously in mixed meals with minimum sample preparation.
Technical Abstract: Near-infrared (NIR) spectroscopy was evaluated as a multiple component analytical technique for prediction of macronutrients and energy in mixed meals with high moisture content. Meals were prepared for spectral analysis by homogenization and NIR spectra (400-2498 nm) obtained with a dispersive NIR spectrometer. Moisture, protein, fat and ash were measured by AOAC methods and carbohydrate and energy were calculated. Using multivariate analysis software, PLS models (n=115) were developed to relate NIR spectra of homogenized meals to the corresponding reference values. The models predicted major components and energy in independent validation samples (n=38) with coefficient of determination of above 0.95. RPD values were between 3.2 and 6.6, and indicated adequacy for screening, quality control, or process control. The prediction accuracy was highest for protein, followed by energy, moisture, carbohydrate, and fat. NIR spectroscopy provides a screening technique for the simultaneous analysis of major components and energy in mixed meals without the need for sample pretreatment beyond homogenization.