Prediction of BMS number using coarseness of marbling particles and shape of M. longissimus thoracis obtained from image analysis for beef rib eye as parameters

Grading of the Beef Marbling Standard (BMS) number would be influenced not only by the ratio of marbling area to M. longissimus thoracis (rib eye) area but also by the coarseness of marbling particles and the shape of the rib eye. The aim of this study was to predict the BMS number with consideration to the coarseness of marbling particles and the shape of rib eye by using the obtained images. The coarseness of marbling particles and the shape of rib eye by computer imaging analysis, which were considered as the possible causes of a decrease in the BMS number, were investigated, and "Coarseness of single particle" was suggested. Pictures of the rib eye area of 446 Japanese Black steers with BMS number assigned by examiners of Wagyu Registry Association were used. The coarseness of a single particle was calculated by dividing the coarseness of the maximum particle by the overall coarseness and was used to distinguish a single particle on the rib eye from other marbling particles. The effect of the coarseness of marbling particles or the shape of rib eye on the grading of the BMS number was investigated to account for the decrease in the BMS number of rib eye with the same level of marbling percentage. For each BMS number, 193 samples that had a larger ratio of fat area than the average ratio were used. The BMS numbers were classified into four levels to examine the cause of the decrease in the BMS number in each level. Next, the BMS number was predicted for 446 carcasses using multiple regressions that considered the coarseness of marbling particles and the shape of rib eye. By using the coarseness of a single particle as index, a single particle on the rib eye that could have caused the decrease in the BMS number could be distinguished. There was a positive and significant correlation between the difference of fat area ratio (the difference between the fat area ratio and the average of the fat area ratio for each BMS number) and the overall coarseness, implying that the overall coarseness would cause a decrease in the BMS number. Also, the effect of the coarseness of marbling particles or the shape of rib eye on the difference of fat area ratio varied in each BMS level. Percentages of the difference between the assigned BMS number and the predicted BMS number being +- 0 and +- 1 were 51.4% and 92.4%, respectively.