Using real-time ultrasound for in vivo estimates of Longissimus dorsi muscle parameters and fat thickness in Dorper ewes

Abstract

Real-time ultrasound has been used as a tool to assist producers to classify sheep into homogeneous groups for slaughter. The aim of this study was to investigate the difference between the in vivo estimates of Longissimus dorsi muscle parameters and fat thickness of mature ewes obtained directly from ultrasound and determined after image analysis with ImageJ. Seventeen Dorper ewes were used, with age of 3.2 ± 1.5 years and body weight (BW) of 59.79 ± 8.81 kg. The estimates of the Longissimus dorsi muscle area (LDA, cm), depth (LDD, cm) and width (LDW, cm) and fat thickness (FT, mm) were performed by ultrasound images (first estimate), after which the images were analyzed using the ImageJ software (second estimate). A randomized design was used with two treatments (US = estimates obtained in vivo by ultrasound; IJ = estimates obtained in vivo by ultrasound and evaluated with ImageJ). The data were submitted to analysis of variance and correlation analysis. Only MW was significantly different between the variables evaluated. No significant correlations were found between BW and the other variables when evaluated by ultrasound and when obtained with the aid of ImageJ. The correlation of LDA (0.91) was high (P < 0.001) in relation to the same estimate from ImageJ. The same pattern was observed for LDDUS and LDDIJ, in which positive correlations of high magnitude (0.95) were also found. The in vivo estimates of LDD and FT of Dorper ewes obtained by ultrasound and those after ImageJ treatment were similar. The high correlations found indicate that the ImageJ program has potential use for the in vivo estimation of Longissimus dorsi muscle parameters and fat thickness in Dorper ewes.