Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 2 de 2
Filtrar
Mais filtros

Base de dados
Ano de publicação
Tipo de documento
País de afiliação
Intervalo de ano de publicação
1.
Meat Sci ; 78(4): 391-9, 2008 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-22062457

RESUMO

Near infrared spectroscopy technology (diode array instrument) was used to study the feasibility of applying quality controls to typical Spanish sausages by performing a proximate analysis (fat, moisture and protein) on the finished product (intact and homogenized). This could be used to provide quality controls at various stages once the finished product was obtained: finished product, storage, distribution and marketing. The selected models were calibrated and evaluated by cross and external validation. For intact products, coefficients of determination for calibration (R(2)) for fat, moisture and protein were 0.98, 0.93 and 0.97, respectively. These values for homogenised products were 0.99, 0.98 and 0.97, respectively. The standard errors of prediction (SEP) for external validation in intact products were 1.47%, 0.97% and 1.08% for fat, moisture and protein, respectively. In homogenised products, these values were lower: 0.71%, 0.41% and 0.95%, respectively.

2.
Meat Sci ; 71(3): 490-7, 2005 Nov.
Artigo em Inglês | MEDLINE | ID: mdl-22060924

RESUMO

Five homogenized meat mixture treatments of Iberian (I) and/or Standard (S) pork were set up. Each treatment was analyzed by NIRS as a fresh product (N=75) and as dry-cured sausage (N=75). Spectra acquisition was carried out using DA 7000 equipment (Perten Instruments), obtaining a total of 750 spectra. Several absorption peaks and bands were selected as the most representative for homogenized dry-cured and fresh sausages. Discriminant analysis and mixture prediction equations were carried out based on the spectral data gathered. The best results using discriminant models were for fresh products, with 98.3% (calibration) and 60% (validation) correct classification. For dry-cured sausages 91.7% (calibration) and 80% (validation) of the samples were correctly classified. Models developed using mixture prediction equations showed SECV=4.7, r(2)=0.98 (calibration) and 73.3% of validation set were correctly classified for the fresh product. These values for dry-cured sausages were SECV=5.9, r(2)=0.99 (calibration) and 93.3% correctly classified for validation.

SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA