A recent research collaboration from China demonstrates the advantages of using hyperspectral imaging to analyze nutritional components in pet food.
In an effort to improve the quality of pet food, a recent research collaboration from Beijing, China, implemented a new approach to analyze the chemical composition of different types of pet food. The findings were published in Microchemical Journal (1).
Hyperspectral imaging (HSI) is an advanced analytical technique used to assess the chemical composition of pet food by capturing detailed spectral information across a wide range of wavelengths (2). This technology allows for the identification and quantification of various components, such as proteins, fats, carbohydrates, and moisture, without the need for physical contact or destructive sampling (2). By analyzing the specific spectral signatures of these components, HSI provides a comprehensive chemical profile of the pet food, ensuring quality control and consistency in formulation. This method can detect contaminants, adulterants, and nutritional imbalances, contributing to safer and healthier pet food products (2). The non-invasive nature of HSI also makes it a valuable tool for real-time monitoring during the manufacturing process, leading to enhanced efficiency and product quality (2).
In this most recent study, researchers from three Chinese institutions (Institute of Quality Standards and Testing Technology for Agro-Products of the Chinese Academy of Agricultural Sciences, Jiangsu Vocational College of Agricultural and Forestry, and Beijing Veterinary Medicine and Feed Control Center) implemented a novel approach to analyze the nutritional quality of pet food (1). This new approach to analyzing the chemical composition of pet food seeks to improve on traditional wet chemical methods, which are time-consuming and labor-intensive. The integration of HSI technology offers a rapid and non-destructive alternative, capable of analyzing various nutritional components such as moisture, crude protein (CP), crude fat (CF), crude fiber (CFe), crude ash (CA), calcium (Ca), and total phosphorus (TP) (1).
The researchers used 36 dog food and 70 cat food samples. Two models were developed using partial least squares regression (PLSR) and tested: one specific to cat food and a mixed model that included both cat and dog food data (1).
The researchers utilized the competitive adaptive reweighted-sampling (CARS) algorithm to select characteristic wavelengths, optimizing the performance of the PLSR models (1). Excluding crude fiber, the cat food model performed similarly to the mixed model when the full spectral bands were used (1). Specifically, the mixed models showed high prediction accuracy for CP, CF, moisture, and CFe, with R2p values ranging from 0.73 to 0.96 and RPD values between 2.22 and 5.20 (1).
The study demonstrated the ability to rapidly and accurately determine the nutritional quality of pet food products using HSI could lead to significant improvements in quality control processes. The researchers used this technology to show the chemical components within the pet food samples, and it allowed them to gain further insight into the quality of the pet food (1). As a result, the researchers demonstrated how their technology could be beneficial to quality assurance teams responsible for ensuring the safety of food products.
In addition, the researchers also demonstrated that their method provides a theoretical and technical foundation for the development of optimized HSI systems tailored for pet food analysis (1). This advancement aligns with the industry's ongoing efforts to ensure the safety and nutritional adequacy of pet food, meeting both regulatory standards and consumer expectations (1).
Although the current study represents a significant leap forward, the researchers mentioned some limitations to their method. They acknowledged the need for further investigation to enhance the accuracy of predictions for certain components like TP, CA, and Ca (1). As a result, their method requires further refinement, which could be explored in a future study.
(1) Xiaolu, L.; Shouxue, L.; Ting, Y.; et al. Predicting the Chemical Composition of Pet Food with Hyperspectral Imaging. Microchem. J. 2024, 203, 110903. DOI: 10.1016/j.microc.2024.110903
(2) Huang, H.; Liu, L.; Ngadi, M. O. Recent Developments in Hyperspectral Imaging for Assessment of Food Quality and Safety. Sensors 2014, 14 (4), 7248–7276. DOI: 10.3390/s140407248