Background: Use of traditional methods for determining meat spoilage is quite laborious and time consuming. Therefore, alternative approaches are needed that can predict the spoilage of meat in a rapid, non-invasive and more elaborative way. In this regard, the spectroscopic techniques have shown their potential for predicting the microbial spoilage of meat-based products. Consequently, the present work was aimed to demonstrate the competence of Fourier transform infrared spectroscopy (FTIR) to detect spoilage in chicken fillets stored under aerobic refrigerated conditions.
Methods: This study was conducted under controlled randomized design (CRD). Chicken samples were stored for 8 days at 4 + 0.5 °C and FTIR spectra were collected at regular intervals (after every 2 days) directly from the sample surface using attenuated total reflectance during the study period. Additionally, total plate count (TPC), Entetobacteriaceae count, pH, CTn (Color transmittance number) color analysis, TVBN (total volatile basic nitrogen) contents, and shear force values were also measured through traditional approaches. FTIR spectral data were interpreted through principal component analysis (PCA) and partial least square (PLS) regression and compared with results of traditional methods for precise estimation of spoilage.
Results: Results of TPC (3.04-8.20 CFU/cm2), Entetobacteriaceae counts (2.39-6.33 CFU/cm2), pH (4.65-7.05), color (57.00-142.00 CTn), TVBN values (6.72-33.60 mg/100 g) and shear force values (8.99-39.23) were measured through traditional methods and compared with FTIR spectral data. Analysis of variance (ANOVA) was applied on data obtained through microbial and quality analyses and results revealed significant changes (P < 0.05) in the values of microbial load and quality parameters of chicken fillets during the storage. FTIR spectra were collected and PCA was applied to illuminate the wavenumbers potentially correlated to the spoilage of meat. PLS regression analysis permitted the estimates of microbial spoilage and quality parameters from the spectra with a fit of R2 = 0.66 for TPC, R2 = 0.52 for Entetobacteriaceae numbers and R2 = 0.56 for TVBN analysis of stored broiler meat.
Discussion: PLS regression was applied for quantitative interpretation of spectra, which allowed estimates of microbial loads on chicken surfaces during the storage period. The results suggest that FTIR spectra retain information regarding the spoilage of poultry meat.
Conclusion: The present work concluded that FTIR spectroscopy coupled with multivariate analysis can be successfully used for quantitative determination of poultry meat spoilage.
Keywords: Analytical methods; FTIR spectroscopy; Meat spoilage; Meat spoilage detection; Multi-variant analysis.