| Curing and thermal processing are two important process conditions which affect both the taste and aroma of meat products.These factors can also act in synergy to prevent the growth of pathogens on meat products.Yet,they also trigger a variety of mechanisms such as lipid oxidation,flavor,color,structure and textural changes in meat which affect consumer’s acceptance of the processed product.The present study developed a new method using electronic nose(e-nose)and hyperspectral imaging(HSI)together with confocal microscopy(CLSM)to explore the lipid oxidation rule of sodium chloride(NaCl)dry-cured pork in different thermal processing times and to predict and visualize changes in the quality characteristics of the processed samples.1.PREDICTION OF QUALITY PARAMETERS OF THERMAL PROCESSED DRY-CURED PORK MEAT USING HSITaking pork belly as a sample to cure with NaCl,the effect of different crystal sizes on the amount of salt absorbed by the pork samples during curing was investigated and the optimal size of the salt crystal for dry-curing was selected.With the aid of the pork curing process based on the optimum crystal size of salt,characteristics of lipid oxidation during the thermal treatment of dry-cured pork were analyzed including variation in moisture content(MC),intramuscular fat(IMF),peroxide value(PV),thiobarbituric acid reactive substances(TBARS),color and texture properties such as hardness,springiness,chewiness and gumminess.And then the feasibility of using HSI to predict selected quality parameters of pork meat was investigated.Optimum wavelengths within 438-955 nm were selected from the spectral and Gray Level Co-occurrence Matrix features to build a combined attribute feature(CAF)model using partial least squares regression(PLSR).This yielded good results with the best determination coefficient of prediction(R_p~2)being 0.896,0.891;and Root Mean Square Error of Prediction(RMSEP)being 5.398,6.180 respectively for hardness and TBARS.The robustness of the CAF model was further tested to predict the MC,IMF,PV,TBARS,color and the texture properties of independent samples.The best validation results(R_p~2=0.829,0.810,0.811,0.811;and RMSEP=2.275,7.686,3.927,1.109)respectively for hardness,chewiness,MC,and TBARS were obtained.2.VISUAL ANALYSIS OF QUALITY PARAMETERS OF THERMAL PROCESSED DRY-CURED PORK MEAT BY HSI AND CLSMApart from the accurate prediction of MC,IMF,PV,and TBARS as quality parameters for the thermal processed dry-cured pork by the CAF model,visualization maps were developed and used to detect their spatial locations in the pork samples.For MC and IMF levels,the distribution maps showed that thermal treatment for 5 hours(O-C-5),10 hours(O-C-10),and15 hours(O-C-15)resulted in high moisture loss and increased IMF content.For TBARS and PV,there was a gradual increase from the cured sample through to the oven cooked samples.In view of the fact that HSI does not provide structural change maps,the CLSM was used for this purpose.The CLSM revealed a number of structural changes,including the collapse of cell compartments and pore openings.These changes may have led to the high moisture loss,hardness and the formation of TBARS and PV recorded in the previous study.3.ANALYSIS OF VOLATILE COMPOUNDS AND OXIDATION ATTRIBUTES OF THERMAL PROCESSED DRY-CURED PORK MEAT BY E-NOSE AND GC-IMSThe breakdown of lipids and fats in cured and thermal processed meat is associated with high levels of TBARS,PV and different volatile organic compounds.In this investigation,the e-nose and gas chromatography coupled with ion mobility spectrometer(GC-IMS)were used to identify the volatile compounds and to predict IMF,PV and TBARS.From the experimental results,35 volatile compounds were identified and quantified by the GC-IMS.Among them,hexanal and pentanal(aldehydes)and 1-pentanol,phenethyl alcohol,1-hexanol and 1-octen-3-ol(alcohols)showed consistent rise in intensities from fresh samples to the cured and cooked samples.Principal component analysis also revealed that apart from the fresh and cured samples which were distinctly different in terms of volatile compositions,those of the O-C-5 to O-C-20 were not so different.For quantitative analysis based on e-nose signals,the prediction result were(Rc~2=0.853;0.805;0.973;RMSEC=2.105;4.616;1.091;R_p~2=0.762;0.771;0.915;RMSEP=1.177;3.136;2.310)for IMF,PV and TBARS respectively.These results led to the conclusion that the TBARS prediction based on the e-nose sensor signals was good,but the IMF and PV predicted results were unsatisfactory.As a result,further steps were taken to improve the accuracy of the prediction of these attributes as outlined in the next section.4.ANALYSIS OF FAT AND PEROXIDE VALUE CONTENT OF THERMAL PROCESSED DRY-CURED PORK MEAT BASED ON DATA FUSION APPROACHBased on the advantages of multi-information fusion,a new approach of detecting IMF and PV of thermal processed dry-cured pork using HSI and e-nose data fusion was proposed in this paper.A PLSR model based on hyperspectral data called CAF was established which yielded R_p~2=0.785,0.840;RMSEP=2.301,1.409 respectively for IMF and PV.A PLSR model based on mean signal values(MSV)from e-nose signals also yielded R_p~2=0.736,0.789;RMSEP=1.450,1.407 respectively for IMF and PV.Finally,data fusion of CAF and MSV was performed which yielded improved prediction results R_p~2=0.841,0.888;RMSEP=1.240,1.816 respectively for IMF and PV.Results from the fusion information suggested that data fusion could provide a viable method for rapidly quantifying IMF and PV evolution of pork meat during processing.The present study used e-nose,HSI technology and CLSM to analyze changes in quality characteristics that occur during processing from fresh pork to bacon-like pork meat products.From the results obtained,new methods and ideas presented in this study for the quality and flavor monitoring and control of pork products could be used to monitor the production process of other congeners or similar foods in order to ensure their safety and quality as well as their consistency in flavor. |
PDF Full Text Request