Visible And Near Infrared Spectroscopy For PSE-Like Zones Classification At Different Post Mortem Times

Thierry Lhommeau (1) , Antoine Vautier (1) , Jean-Marc Goujon (2) , Ronan Le Page (2)

1 IFIP - Institut du Porc
2 FOTON - Fonctions Optiques pour les Technologies de l'informatiON

– The ability of visible and near infrared spectroscopy (VIS-NIRS) to predict PSE-like zone classification was examined on 150 deboned hams at 5 post mortem times (12, 16, 20, 24 and 36 h pm). Four probes were used to investigate the accuracy. Two of our probes were prototypes. Partial Least Squares Discriminant Analysis (PLSDA) was used to determine the prediction models from spectrums recorded on semimembranosus muscle. Surface probes gave better results than inserted probes. The best probe showed moreover a good accuracy whatever the pm times was when measuring (78 to 86 % good prediction in external validation as the 24 h pm model was used). Its false negative/false positive ratios were favorable at the early pm times (12 to 20 h pm). So an early PSE-like zones prediction is possible using VIS-NIRS. The prototype probe inserted showed good results (82,9% good prediction in external validation) and would deserve additional validation on bone-in hams, as industrials need. Key Words – VIS-NIR spectroscopy, prototype probes, pork meat, PSE-like zones, PLSDA.

Domaines: Alimentation et Nutrition; Optique / photonique

Thierry Lhommeau, Antoine Vautier, Jean-Marc Goujon, Ronan Le Page. Visible And Near Infrared Spectroscopy For PSE-Like Zones Classification At Different Post Mortem Times. 61st International Congress of Meat Science & Technology (ICOMST 2015), Aug 2015, Clermont-Ferrand, France. ⟨hal-01234315⟩