The innovative model enables fast, non-invasive prediction of four freshness indicators, providing a promising tool for enhancing food safety in the fish industry.
Bighead carp, a prominent species in the aquaculture industry, is highly susceptible to rapid deterioration due to its high protein and moisture content. Microbial growth accelerates post-mortem, affecting the fish’s freshness. Monitoring temperature fluctuations in cold chains – a crucial aspect of fish transport and storage – is vital for maintaining quality. Existing methods for determining fish freshness, including various spectroscopic techniques, offer great potential but often lack precision in real-time monitoring. This study combines EEM fluorescence spectroscopy with LSTM to create a more accurate predictive model for freshness assessment in simulated cold chain environments.
A study (DOI:10.48130/fia-0024-0037) published in Food Innovation and Advances on 13 December 2024 by Ce Shi’s team, Beijing Academy of Agriculture and Forestry Sciences, offers a non-destructive and real-time solution for freshness monitoring of bighead carp in cold chains, providing significant improvements in food safety and quality assurance.
The study employed a combination of chemical analysis and excitation-emission matrix (EEM) spectroscopy to assess the freshness of bighead carp heads during storage under different temperature conditions. The chemical analysis included monitoring key indicators like TVB-N (total volatile basic nitrogen), TBARS (thiobarbituric acid reactive substances), K value, and TVC (total viable count), which reflect protein degradation, fat oxidation, and microbial growth. Results showed that TVB-N, TBARS, K value, and TVC all increased with storage time, with the highest rates observed at 16 °C. Specifically, the TVB-N value exceeded 20 mg/100 g, indicating spoilage at 16 °C after 3 days, while TBARS values remained under 0.60 mg·kg−1, with higher storage temperatures showing slower oxidation rates. Similarly, the K value, a freshness indicator, peaked rapidly at 16 °C, signifying decomposition. The TVC values surpassed the threshold of 7.00 log10 CFU/g, indicating microbial spoilage. The EEM analysis, using the PARAFAC method, identified three key fluorescent components linked to tryptophan and NADH, with fluorescence intensity increasing during storage, particularly under higher temperatures. The fluorescence data were then integrated into an LSTM (long short-term memory) model to predict freshness. The EEM-LSTM model demonstrated excellent performance, with high R² values and low RMSE, MAE, and MAPE across all freshness indicators. The model successfully predicted freshness parameters such as TVB-N, TBARS, K value, and TVC, with errors lower in the supermarket direct sales (SDS) cold chain compared to the long-distance transport (LDT) chain. These findings suggest that combining EEM spectroscopy with deep learning models offers a powerful, non-destructive method for real-time freshness monitoring of aquatic products under varying storage conditions.
The integration of EEM fluorescence spectroscopy with LSTM deep learning represents a significant leap forward in the non-invasive monitoring of seafood freshness. This study not only paves the way for more accurate and efficient quality control in the fish industry but also sets the stage for future innovations in the freshness prediction of perishable goods within cold chains.
###
References
DOI
10.48130/fia-0024-0037
Original Source URL
https://doi.org/10.48130/fia-0024-0037
Funding information
This study was supported by the Beijing Academy of Agriculture and Forestry Sciences Outstanding young scientist training program, the Fund of Young Beijing Scholar, Jiangsu Science and Technology Plan (Key Research and Development Plan Modern Agriculture) Project (BE2023315), and the Ministry of Finance and Ministry of Agriculture and Rural Affairs: National System of Modern Agricultural Industry Technology (CARS-45-28).
About Food Innovation and Advances
Food is essential to life and relevant to human health. The rapidly increasing global population presents a major challenge to supply abundant, safe, and healthy food into the future. The open access journal Food Innovation and Advances (e-ISSN 2836-774X), published by Maximum Academic Press in association with China Agricultural University, Zhejiang University and Shenyang Agricultural University, publishes high-quality research results related to innovations and advances in food science and technology. The journal will strive to contribute to food sustainability in the present and future.