Skip to main content
ARS Home » Research » Publications at this Location » Publication #149174

Title: MEASUREMENT OF DIETARY FIBER IN MIXED FOODS USING NEAR INFRARED SPECTROSCOPY (NIRS)

Author
item Singh, Mukti
item Kays, Sandra

Submitted to: Eastern Analytical Symposium
Publication Type: Abstract Only
Publication Acceptance Date: 6/16/2003
Publication Date: 11/19/2003
Citation: Singh, M., Kays, S.E. 2003. Measurement of Dietary Fiber in Mixed Foods Using Near Infrared Spectroscopy (NIRS). Eastern Analytical Symposium. Abstract #343; p. 37.

Interpretive Summary:

Technical Abstract: The Nutrition Labeling and Education Act of 1990 requires food processors to report total dietary fiber (TDF) on consumer product labels. TDF is currently estimated by the long and laborious AOAC method, however, NIRS has been successfully used to measure TDF in cereal products. An attempt was made to measure dietary fiber in mixed food using NIRS. Products selected for the study were commercially available heat-to-eat dinners and foods, having TDF in the range of 1-12 percent for both the calibration and independent validation sets. They were blended as is from the package in the Robot Coupe food processor. The blended samples were dried in the hot air oven and then defatted using Soxhlet apparatus. The dried samples were ground with a coffee grinder and the defatted samples in a Satake rice mill. The samples were scanned 1) as is (freshly bended), 2) dried and ground, and 3) defatted and ground using NIRSystem 6500 (400 - 2500 nm) and Bruker Vector 22/N-FT (10000 - 4000 cm-1). TDF was measured by the enzymatic-gravimetric AOAC method (991.43) on the dried defatted samples. Commercial chemometrics packages, WinISI, and Unscrambler were used to develop calibrations based on the scans and assay values for TDF. The calibration models developed were evaluated based on the coefficient of determination and standard error of cross validation for the calibration set and coefficient of determination, standard error of prediction, slope, bias and rpd for the validation data set.