Electronic nose-based monitoring of vacuum-packaged chicken meat freshness in room and refrigerated storage

Monitoring chicken meat is a crucial process for food safety and consumer health, as it helps prevent the growth of harmful bacteria, minimizing the risk of foodborne illnesses. Currently, electronic nose (E-nose) technology plays a sig-nificant role in food quality assessment as it can detect changes in volatile compounds associated with food freshness. In this study, a self-designed, cost-efficient E-nose system was introduced to evaluate the freshness and bacterial growth of vacuum-packaged chicken meat stored at room temperature and refrigerator at 4 degrees C. Polynomial feature extraction with varying degrees was employed to extract important information from the sensor responses. Principal component analysis (PCA) and linear discriminant analysis (LDA) were implemented for data dimensionality reduction and classification. A support vector regression (SVR) model was built and employed to evaluate the bacteria population based on response pat-terns from the E-nose device. The LDA results clearly showed the classification of chicken meat freshness corresponding to different storage days and temperatures. The E-nose device with the SVR model combined with extracted parameters using a 2-degree polynomial provided good prediction results for the bacteria population with high R-T(2) scores of 0.99 and 0.99, R-CV(2) scores of 0.97 and 0.93, and minimum RMSET scores of 0.08 and 0.11 log Cfu/g, and RMSE(CV )scores of 0.24 and 0.30 log Cfu/g, at room and refrigerated temperatures, respectively. The results indicate that the developed E-nose system could be used as a fast, portable, low-cost, and non-destructive measurement tool in evaluating the bacterial growth of chicken meat with high relative accuracy