Measuring intramuscular fat in beef with ultrasonic frequency analysis

Area of Science:

• Food Science
• Biophysics
• Meat Science

Background:

• Intramuscular fat (IMF) content is crucial for beef quality and consumer preference.
• Accurate prediction of IMF is essential for grading and processing.
• Non-invasive methods for IMF assessment are highly desirable.

Purpose of the Study:

• To evaluate the efficacy of ultrasonic Fourier spectral analysis for predicting IMF content in beef.
• To identify key frequency-domain parameters for IMF prediction.
• To develop and validate a multivariate model for IMF estimation.

Main Methods:

• Ultrasonic signals were analyzed using Fourier transform to obtain frequency spectra.
• Correlation analysis was performed to identify significant spectral parameters.
• A multivariate regression model was developed using frequency-domain parameters.
• The model's predictive capability was validated on beef tissue samples.

Main Results:

• The number of local maxima in the Fourier spectrum was the most significant predictor of IMF concentration (r = .89).
• An optimum frequency of 1.92 MHz was identified for IMF prediction.
• A multivariate model (R² = .82) accurately predicted IMF, with an average error of 1.17%.
• The model demonstrated a mean accuracy of approximately 79% and was most effective for IMF below 4%.

Conclusions:

• Ultrasonic Fourier spectral analysis is a viable non-invasive method for predicting beef IMF content.
• Frequency-domain parameters, particularly local maxima, offer valuable insights into fat distribution.
• The developed multivariate model provides a reliable tool for estimating IMF, especially in lower ranges.