[Rapid evaluation of beef quality by NIRS technology]

Area of Science:

• Food Science and Technology
• Analytical Chemistry
• Spectroscopy

Background:

• Traditional methods for assessing beef quality are often time-consuming and labor-intensive.
• Objective and rapid evaluation of beef quality is crucial for the meat industry.

Purpose of the Study:

• To develop a rapid near-infrared reflectance (NIR) spectroscopy method for evaluating beef quality.
• To establish a Partial Least Squares (PLS) prediction model for key physic-chemical characteristics of beef.

Main Methods:

• Collected 114 beef samples from five different carcass parts after 48 hours of aging.
• Acquired spectra using NIR spectroscopy (950-1650 nm) and applied preprocessing techniques (MSC, SNV, first derivative).
• Developed PLS models to predict moisture, fat, protein, pH, color (L*, a*, b*), and Warner-Bratzler Shear Force (WBSF).

Main Results:

• Achieved high predictive correlation coefficients for moisture (0.9472), fat (0.9245), and protein (0.9346).
• Obtained moderate to good predictions for color parameters (L*: 0.8203, a*: 0.8646, b*: 0.7530).
• External validation with 30 additional samples confirmed no significant difference between predicted and conventional laboratory values.

Conclusions:

• Near-infrared reflectance (NIR) spectroscopy is a rapid, effective, and accurate technique for evaluating beef quality.
• The developed PLS models demonstrate good predictive ability for both chemical and physical characteristics.
• Chemical characteristic predictions showed higher accuracy compared to physical characteristic predictions.