Prediction of Total Dietary Fiber in Cereal Products Using Near-Infrared Reflectance Spectroscopy

Abstract

The Nutritional Labeling and Education Act of 1990 has increased the need for more rapid, efficient, and nonpolluting techniques of analyzing the nutrients in foods, particularly dietary fiber. The use of near-infrared spectroscopy (NIRS) for the prediction of total dietary fiber content of cereal products was investigated in this study. Cereal and grain products, including breakfast cereals, flours, brans, crackers, and samples containing commercial oat and wheat fibers, were selected for analysis. Products (n = 91) were dry milled, and total dietary fiber was measured by the AOAC (991.43) enzymatic-gravimetric method. Total dietary fiber values ranged from <1 to 52% of dry weight. Milled cereal products (n = 91) were scanned from 1100 to 2498 nm with a NIRSystems 6500 monochromator. Using ISI software for scanning and data analysis, a dietary fiber calibration was obtained with partial least squares as the regression method. The standard error of cross validation and multiple coefficient of determination were 1.58% and 0.99, respectively. For equation validation, an independent group of cereal products (n = 31) was dry milled, and total dietary fiber was determined. The samples were scanned, and total dietary fiber was predicted by NIRS. Samples were predicted with a standard error of performance of 1.51%, coefficient of determination of 0.99, bias of −0.38, and slope of 1.06. This study shows that NIRS can be used to rapidly and accurately predict total dietary fiber content in a wide range of cereal products. Keywords: Dietary fiber; near-infrared spectroscopy; nutrition labeling