The aim of this study was to investigate the meat sample preparation methods and NIRS methodology to predict sensory scores of veal belonging to two ethnic groups and fed on different diets. Three preparations of longissimus thoracis samples, i.e., raw (RA), ethanol-prepared (ET), and freeze-dried (FD), were studied. Thirty-two male calves, 16 Friesian (F) and 16 Crossbreeds (C), were fed milk replacer, maize silage and 65 kg/calf (L) or 100 kg/calf (H) of maize grain. The meat analyses were: water, protein, fat, haem iron content, drip and cooking losses, colour, Warner-Bratzler shear (LAB), fatty acids profile (FA) and sensory evaluation on four attributes (Panel). The samples were scanned by a LabSpec-Pro (ASD) portable (UV-Vis-NIRS: 350-2500 nm), the FD samples were also analysed by an electronic nose. Chemometrics MPLS of the NIRS spectra and of LAB and FA were performed to get distance matrices between groups and prediction performances of Panel scores. The matrices reached different R2 levels: 0.65 (RA); 0.65 (ET); 0.62 (Panel); 0.42 (FD); 0.42 (LAB); 0.04 (FA) and 0.50 (E-nose of FD). Clusters from NIRS of raw samples corresponded to the same pattern obtained by LAB and FA. Prediction of Panel scores from the 32 veal calves were effective as R2 cross-validation of ET specimens were: Visual 0.11, Flavour 0.68, Texture 0.68 and Global 0.53. It was concluded that NIRS scan of RA samples anticipates results achieved by a wide set of laboratory analyses. NIRS analysis of ET samples exhibited strong predictive value of Panel scores.
NIRS analysis of different meat sample preparations from veal calves and panel test prediction
BRUGIAPAGLIA, Alberto;DESTEFANIS, Gianluigi;LUSSIANA, Carola;
2011-01-01
Abstract
The aim of this study was to investigate the meat sample preparation methods and NIRS methodology to predict sensory scores of veal belonging to two ethnic groups and fed on different diets. Three preparations of longissimus thoracis samples, i.e., raw (RA), ethanol-prepared (ET), and freeze-dried (FD), were studied. Thirty-two male calves, 16 Friesian (F) and 16 Crossbreeds (C), were fed milk replacer, maize silage and 65 kg/calf (L) or 100 kg/calf (H) of maize grain. The meat analyses were: water, protein, fat, haem iron content, drip and cooking losses, colour, Warner-Bratzler shear (LAB), fatty acids profile (FA) and sensory evaluation on four attributes (Panel). The samples were scanned by a LabSpec-Pro (ASD) portable (UV-Vis-NIRS: 350-2500 nm), the FD samples were also analysed by an electronic nose. Chemometrics MPLS of the NIRS spectra and of LAB and FA were performed to get distance matrices between groups and prediction performances of Panel scores. The matrices reached different R2 levels: 0.65 (RA); 0.65 (ET); 0.62 (Panel); 0.42 (FD); 0.42 (LAB); 0.04 (FA) and 0.50 (E-nose of FD). Clusters from NIRS of raw samples corresponded to the same pattern obtained by LAB and FA. Prediction of Panel scores from the 32 veal calves were effective as R2 cross-validation of ET specimens were: Visual 0.11, Flavour 0.68, Texture 0.68 and Global 0.53. It was concluded that NIRS scan of RA samples anticipates results achieved by a wide set of laboratory analyses. NIRS analysis of ET samples exhibited strong predictive value of Panel scores.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
