Beef color standard including marbled Wagyu beef determined using a portable near ‐infrared optical fiber spectrometer

Objective: This study determined whether a portable near-infrared optical fiber spectrometer operating within a wavelength range of 700-1,050 nm could objectively predict the Beef Color Standard (BCS) visually evaluated during beef carcasses grading. Methods: A total of 1,077 carcasses were procured from Japanese black (JB) Wagyu (n = 534), crossbred (F1) (n = 367), and dairy cows (n = 176) shipped to meat markets. The cut surface between the sixth and seventh ribs was visually evaluated by grading members using the BCS. Concomitantly, a near-infrared optical fiber spectrometer was applied to the cut surface to obtain the reflectance spectrum. Using data from 827 carcasses, a calibration model was established through a partial least squares regression analysis using the BCS value as the objective variable and the optical data following quadratic differential processing as the explanatory variable. Using data from 250 additional animals, the calibration model and predicted BCS values were verified for accuracy. Results: The calibration model exhibited a high correlation coefficient of 0.83 and a standard error of calibration (SEC) of 0.70. The predicted BCS by near-infrared evaluation rarely differed significantly from the actual BCS, and the difference was roughly distributed in the range of ±1. For verifying the accuracy of our calibration model, the correlation coefficient between the actual BCS and the predicted value for the validation sample was 0.75, and the standard error of prediction was 0.71. The probability that the error between the measured and optically predicted BCS values in the validation samples was within ±1 and ±0 was 97.6% and 54.4%, respectively. Conclusion: Using a portable near-infrared fiber optic spectrometer, we demonstrated that the BCS value of beef carcasses, including marbled Wagyu beef in meat markets, can be objectively evaluated with relatively high accuracy.