Abstract

Measurement accuracy for ultrasonic tissue depths was examined in order to determine the relative importance of factors affecting it. Repeatabilities for ultrasonic fat depth (UFD) and ultrasonic muscle depth (UMD) were assessed from repeated runs (2) in three groups of animals; Dorset Down ewes (LW=45.4kg, 7.5mo, n=50), Coopworth rams (LW=51.5kg, 11mo, n-50) and Dorset Down rams (LW=38kg, 4.5mo, n=50). Repeatability values were higher for UMD (0.77, 0.95 & 0.87 for the 3 groups, respectively) than for UFD (0.63, 0.84 & 0.82, respectively). Lower repeatabilities were due to high within-animal variance rather than low between-animal variance. Measurements made on videotaped images demonstrated that image registration had a great effect on repeatability, with image interpretation having a negligible effect. A comparison of two operators differing in experience showed that fat depth was measured equally accurately by both operators but that the experienced operator measured muscle depth more accurately. It was concluded that tissue depths can be accurately assessed from one measurement.

RE, McDowell

Proceedings of the New Zealand Society of Animal Production, Volume 52, , 3-6, 1992