Impact of freeze-thaw cycles on raw beef quality deterioration and quality prediction using Hidden Markov model analysis

Abstract

The impact of freezing time on the quality of raw beef was investigated in this study. The essential components, thiobarbituric acid reactive substances (TBARS) values, and protein carbonyl content were monitored during freeze-thaw cycles. There were obvious protein and lipid oxidation during the freeze-thaw cycles. The carbonyl content and TBARS values increase from 1.90 ± 0.01 mmol/mg protein and 0.23 ± 0.01 mg/kg of the control group to 2.27 ± 0.03 mmol/mg protein and 0.51 ± 0.02 mg/kg at 7 freeze-thaw cycles. The findings confirmed that freeze-thaw cycles can serve as an acceleration method for freezing storage. The acceleration model was reliable for up to 3 freeze-thaw cycles, with one freeze-thaw cycle being equivalent to 30 days of freezing storage. However, similarity to actual freezing storage decreased rapidly after 3 freeze-thaw cycles. The Hidden Markov Model (HMM) successfully classified and predicted quality changes in raw beef during freezing storage. It accurately predicted the quality level at specific time interval and determined shelf life by tracking the transition from fresh to spoiled state. Overall, HMM proved to be an effective tool for predicting quality changes in raw beef during freezing storage, promoting the maximum utilization of freezing beef.