Repeatability of Ultrasound-Predicted Percentage Intramuscular Fat

Repeatability of ultrasound-predicted percentage intramuscular fat was studied using observations from 144 head of cattle. Animals were scanned at an average age of 433-d by a certified technician. Each animal was scanned five times using two Aloka 500V machines. Percent intramuscular fat was then predicted by placing a box at two different positions within an image. Overall standard deviation of observations within an image was .5% and the within animal standard deviation of observations was estimated at .9%. The overall repeatability was .69. There were slight differences in repeatability when data were analyzed by machine (probe). Prediction of percentage intramuscular fat was more repeatable when boxes were placed in a “best” position rather than specifically between the 12 and 13 ribs. Evaluation of the effect of repeated measurements indicates that increasing the number of images per animal plays a significant role in reducing the standard error of prediction more so than taking more measurements within a single image.