Featured Articles

Interpretive Summary: Using near-infrared spectroscopy to determine intramuscular fat and fatty acids of beef applying different prediction approaches

By Anne Zinn

Beef consumers are often interested in knowing the nutritional content of the product they eat, specifically the amount of fat and description of fat information, which can contribute to a beef purchase decision. It has been determined through various studies that the average consumer is unaware of the fat and fatty acid profile and its nutritional functions, which suggests the need to offer information on fat and fatty acid profiles as part of the meat information label.  With this in mind, there is a growing need for advanced techniques for determining the nutritional components of meat, which can be costly, time consuming, and have negative environmental impacts. A study recently published in the Journal of Animal Science aimed to compare conventional methods and techniques based on machine learning to predict fat content and fatty acid profiles in beef using near-infrared methods.

Results of this study indicated that the application of a combination of wavelength selection through genetic algorithm and regression-based vectors of machine support generated a reliable model with the ability to predict fat and fatty acids. Characterization of the fatty acid profile demonstrated a baseline on which the participation of beneficial fatty acid and the polyunsaturated to saturated fatty acids ratio  in beef can be improved under the extrapolation capacity of the present study. The results could support the development of differentiated beef with a higher capacity for competition and access by consumers concerned about the consumption of healthy foods. The models identified also offer a possibility of providing information for the consumer through a brand that categorizes beef and nutritional information labels based on fat and fatty acid necessary to make an informed purchasing decision. While the models identified here need additional validation, they begin to offer the meat industry a faster and more reliable quantification tool for determining beef fat and fatty acids.

The full paper can be found on the Journal of Animal Science website.