Prediction of beef tenderness using hyperspectral imaging

Cuts from different types of beef muscles [e.g., infraspinatus (Top blade, TB), gluteus medius (Top sirloin, TS), psoas major (Tenderloin, TL), and longissimus thorasis (Rib eye, RE)] were dry-aged for up to 21 days and then subjected to near-infrared hyperspectral imaging to gauge its usefulness in evaluating beef tenderness. Imaged in reflection mode using a hyperspectral (900 nm<  <1700 nm) imaging system, samples were then cooked (grilled one side to internal temperature of 40ºC, turned and grilled to a final internal temperature of 71ºC) and examined for tenderness by the Warner-Bratzler shear force (WBSF) method. Stepwise regression of mean spectral data collected from the samples was used to determine wavebands that can be used assess beef tenderness. Multiple Linear Regression (MLR) was used to assess the relative advantage of the selected wavebands to predict tenderness of beef samples. An overall correlation coefficient (R) for each muscle type (R=0.89 for TS, R=0.86 for RE, R=0.81 for TB, and R=0.83 for TL) shows the possibility of using hyperspectral imaging for predicting beef tenderness.Différentes coupes de muscle de bœuf [e.g., infraspinatus (lame supérieure, LS), gluteus medius (haut de surlonge, HS), psoas (Filet, FL), et longissimus thorasis (entrecôte, EC)] furent âgées à sec pour 21 jours, puis soumises à une imagerie hyperspectrale en infrarouge proche pour évaluer l'utilité de telles mesures dans l'évaluation de la tendreté du bœuf cuit. Des images d'échantillons crus furent prises avec un système d'imagerie hyperspectral (900 nm < 