Effect of Cereal Product Residual Moisture Content on Total Dietary Fiber Determined by Near-Infrared Reflectance Spectroscopy

Abstract

Published studies on the applications of near-infrared reflectance spectroscopy (NIRS) to the analysis of fiber in forages, feeds, grains, and cereal products indicate the presence of O-H absorbance, due to sample moisture content, in the calibration models. The objective of this study was to determine the extent to which residual moisture in samples interferes with the ability of NIRS to predict total dietary fiber (TDF) in cereal products and grains. Milled cereal products and grains were stored in 20%, 60%, and 80% experimental relative humidity (rh) environments and a vacuum oven. Samples (N = 143) were analyzed for moisture and predicted for TDF. Results showed significant differences between laboratory reference TDF and predicted TDF for samples that were either very low or very high in moisture. Cereal products and grain samples stored under ambient conditions (N = 90) were combined with selected samples stored under different rh environments (N = 53) to develop a new calibration using partial least squares regression. The standard error of cross validation and multiple coefficient of determination (R2) were 1.85% and 0.98, respectively. The model was validated with an independent set of cereal products (N = 29) stored under ambient and rh environments. Samples stored under ambient and rh environments were predicted with standard errors of performance of 1.70% and 1.86%, respectively. The study shows that NIRS can be used to predict TDF in cereal products and grains with a wide range of residual moistures when calibrations include the range of residual moisture expected. Keywords: Dietary fiber; near-infrared spectroscopy; moisture