Ability of Near-Infrared Reflectance Spectroscopy to Predict Fat, Cholesterol, and Caloric Content of Fresh and Cooked Ground Beef

摘要

The objective of this study was to quantify fat, calories, and cholesterol content in fresh and cooked ground beef using near-infrared (NIR) spectroscopy. Weighed portions of USDA Select, peeled knuckles and 50:50 trim were ground, mixed, and formed into 0.33-lb patties with an approximate fat content of 5, 10, 15, 20, 25, 30, 35, and 40% (10 patties/fat level/replicate). Near-infrared reflectance (350 to 1,050 nm) was measured on the surface of patties from each fat group. Then, one-half of the patties were pan-fried to an internal temperature of 160 deg F, split longitudinally, and the inner surfaces of patties were scanned with the spectroradiometer. After scanning, fresh and cooked patties were freeze-dried, pulverized, and analyzed for fat, cholesterol, and caloric content. Partial least squares (PLS) regression analysis of the spectrum resulted in validation R~2 values (R_(Val)) for fat, calories, and cholesterol ranging from 0.91 to 0.98 for fresh ground beef and 0.73 to 0.93 for cooked ground beef. Overall predictability for the NIR region of the model had correlation coefficients ranging from 0.92 to 0.99 for fresh beef, and 0.78 to 0.96 in the cooked category. This study indicated that NIR spectroscopy could be a useful tool for qualitymonitoring in the beef industry.